• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Computer reboot when loading a save game

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

jebediah

Member
Joined
Apr 14, 2012
Location
Bitter cold Sweden
Well here's a tricky one i can't solve by my own.
I'm placing it nVidia section since i think it is the GPU causing the problem.

I just gave my brothers son my old computer.
The specs is
*i7 930
*Corsair Dominator DHX+ DDR3 1600MHz 6GB
*ASUS P6T Deluxe V2, Socket-1366
*MSI GTX 760
*Thermaltake Toughpower 750w PSU
*1TB WD HDD

When i used this computer it run just fine with my two HD7970 in CF.

To the problem
It's two scenario.
*Every time he/i try to load a saved game, the computer just reboot. No error message, no BSOD and nothing in the event viewer.
*When we launch a game it seems to run perfectly. You can change all the settings etc, but just a secong into the intro in the game. Same reboot appear from no where.

And it seems like it happens more often in "easier" games like LEGO and older games.



I have tried almost everything now.
*First, all drivers/bios are up to date.
*I've tried older driver for the video card. Back to 320.49.
*I've tried Turbo boost/Speedstep/and all C-states both on and off.
*I've tried both stock CPU speed and a slight OC
*Prime95 gives me around 75c as highest on one CPU core. Did make a run over the night.
*memtest+ can't find any errors. Did make a run over the night.
*OCCT can't find any errors what so ever.
*Both 3DMark and UNIGINE's Heaven runs just fins. UNIGINE's Heaven was looping over the night. The highest GPU temp stayed around 77-78c


So there's only two things a suspect from this point.
*The PSU? But OCCT PSU benchmark can't find any problems.
*The GPU. Why? I'm not 100% sure since the reboot happens to quick and suddenly. But right before the reboot happens "and remember, it's always happens when the GPU usage drops" the GPU core voltage drops down to 0.8, the fans make a little less noise, and a ½ sec after that, boom.


Since i'm a AMD user my self i have no experience with nVidia. Is there any power saving function in the registry you can change?
I've already change it in the nVidia CP.
 
Last edited:
What OS? Although it could be hardware related, another common denominator is your OS launching the games. Have there been any updates to the OS or add-ons such as virus protection, safety/security settings, power saving features on/off that could also affect the system?
 
Win7. I've been using the same "setup" for several years.
The computer is formated and a clean install of Win. Only the necessary updates.
And only standard programs like Winrar, Mediaplayer+codec, Microsoft essential etc.

Gonna give it a try later today and uninstall some programs.
 
I don't believe this to be a PSU or PCIe bus issue... if so, it would be happening when gaming or stress testing.

Im sorry I do not have any other suggestions...
 
I overlooked that last line re: Unigine. Perhaps I'm mistaken. Though any SW issues should result in a BSOD. A flat out power off/reset is usually power related.
 
Indeed, just curious that when he uses the most power, it doesn't have a problem.

Can you use a prior save game and it still shows the same behavior? And this is on ONE game only correct? Something seems up with the save game file or the game to me if it is the only one having a problem.
 
Hi again. Back from the work so it's time to dig deeper in the problem.

Again, it's not my computer, not my games and not me who are playing the games.
But the little guys told me that he had problems with Lego superheroes marvel, Tomb raider, Minecraft and Batman arkham asylum.

He told me that sometimes "2 out of 5 time" he was able to start a new game, and where able to play some minutes before the reboot. And sometimes the reboot came directly before/after the intro "as i described earlier"

My first opinion when i saw the problem was maybe that some game did not like when GPU downclocked it self while gaming.
And i have no idea if there's any registry tweaks for that.

There is some good news. Now i'm able to start games again and play, but i don't know for how long. I guess the problem appears again right after i deliver the computer to him.

Also forgot to mention that i did a run with HDD regenerator. No bad sector was found.

Will update the thread later on.


EDIT: 1
Here's some numbers from OCCT GPU Benchmark
1.72 V - WIN1
3.29 V - +3.3V
4.92 V - +5V
11.87 V -+12V
1.40 V - VIN6
0.89V - VIN0
 
Last edited:
im curious as to how this turned out as i am having a very similar issue:

had a sabertooth z77 for a few years, ran it with a GTX 670 with no issues, i bought a gtx 780 and the PC started to randomly reboot. no errors, just shut off like i pulled the power plug.

i thought, ok, maybe power supply- i have an OCZ 1000 watt gold rated that i got when i picked up the sabertooth, i thought maybe the additional power draw stressed a weakness in PSU, so i went and bought EVGA Supernova 750G and all was good for like a week, then same thing, power cycles randomly. must be a bad card... returned it an bought a ASUS GTX780 ti Directvcu ii OC and installed... same thing. i do some googling and see some issues with "dirty power" in the house so i go buy a UPS rated to 700 watt that claims a feature that "conditions" the power to account for drops/surges in the power and it still happened.

i installed an old gtx 560ti DS and it still happens... saw something "anti surge" so i disabled it and still reboots..whats left?

then i found that i could "trigger" the reboot. Seems if i run 3DMark, ice storm demo it will reboot every time. i also noticed that when the demo is running, before it shuts off, i can hear a weird whine (?)i don't know what to call it but sounds bad, i cant really pinpoint it, but it seems to be coming from the north bridge part of the board so i thought the sabertooth had an issue. i went and got an ASUS Maximus VI Hero and a new CPU (i7 4770k) and ran the "ice storm demo" and reboot with that ugly whine noise...

i dont know what to do next... what to check. i guess haven't bought new RAM yet, but i did run memtest from a USB (outside of windows) overnight and all passed, no errors. heres the kicker... i can run firestrike fine. other benchmarks like heaven on extreme setting and OC scanner fine. but that ice storm demo, the test is for low end laptops and tablets and stuff.. i just dont know what else to try, could it be a drive?

I have (all stock fresh install);
ASUS Hero mobo
Intel i7 4770k
Corsair CMD8GX3M2A 1866C9 (currently running at default asus bios settings)
ASUS gtx780 ti Directcu ii
SSD OCZ-VERTEX4 256GB (primary drive with OS)
SSD OCZ-AGILITY3 128GB (game drive)
3 HDD 1TB (Storage)
OCZ Fatality 1000watt modular PSU
 
I'd start by making sure nothing is causing a short in any of the mounting.

Does event viewer show anything happening after the restarts?

Sounds wonky, but try a different mouse and keyboard. Sometimes the drivers can cause issues that don't show up as conflicts (OpenCL, .Net framework, etc)

Disconnect optical drives, and any unnecessary USB peripherals etc.

If you can use/borrow a stick of RAM to rule that out it wouldn't hurt....
 
I still wonder if it's the PSU since it is not BSOD and it's rebooting or something is overheating, is all your fans working?
 
I still wonder if it's the PSU since it is not BSOD and it's rebooting or something is overheating, is all your fans working?

The OP could very well find a solution therein

Mrcold I wouldn't be so sure about. 2 decent PSU's failing? One of which happening within a week of purchase.. possible, and it's happened before but.... unless something else is tripping OCP or something..
 
The OP could very well find a solution therein

Mrcold I wouldn't be so sure about. 2 decent PSU's failing? One of which happening within a week of purchase.. possible, and it's happened before but.... unless something else is tripping OCP or something..

OCP<-- over current protection? is this a feature of the PSU? I found a setting for Anit surge in the BIOS for both the sabertooth and the HERO, i disabled it in the sabertooth, ive left it alone on the hero.

im looking @ OCCT 4.4.0 voltage monitoring and the 12v rail is 12.1 constant while idle, ran a quick bench and it went from 12.09 to 12.00, i attached a SS
 

Attachments

  • 2014-04-15-01h23-Voltage-+12V.png
    2014-04-15-01h23-Voltage-+12V.png
    45.3 KB · Views: 50
New Information

Ok, so i pulled the 780 out and ran with the integrated i7 gfx and no whining noise and no reboot. how could there be a problem with discrete graphics on 2 different mobos with three different gfx cards?

could the mobo be shorting out somehow... might it be the case?

i have a corsair 600T... what should i look for? i have screws in all but one of the nine stand offs (one in the middle of the mobo area is just like a place holder post, the top pokes through the whole but you cant screw anything into it.

before i pulled the 780 i tested the RAM one stick at a time, disconnected all but the OS SSD drive and even installed windows on a backup ssd.

could it be there is a short of some kind messing with the PCIe lanes/controller/whatever?
 
Well the problems is still not solved.
I'm gonna dig deeper into the problem next week starting with the HDD.

I've done everything possible for testing "prime95, 3DMark, memtest86+ etc" without any issue and the problem still only occurs when my brothers kid are playing.

As some has already stated, it starting to look as it should be a ground fault.
I wish i had a extra GPU for testing. A have a X1900XTX lying around somewhere. Just need to find it.
 
Back