Sign in with
Sign up | Sign in
Your question
Solved

New System; Game Programs Freeze/Become Unresponsive Always After Short Time Played

Last response: in Systems
Share
September 6, 2013 5:36:15 PM

Hello, I've been having a pretty specific and unusual problem with a new computer that I built for gaming.

System Specs:
-Windows 7 Ultimate
-Intel Core i5-3570K @ 3.4GHz
-ASUS P8Z77-V Z77 Mobo
-MSI GeForce GTX 660 Twin Frozr 2GB GDDR5
-Patriot 8GB Viper Xtrm Div2 DDR3 1600MHz
-Ultra 750W Modular PSU (from 2007; the only old component of the system)
-Cooler Master Hyper N520 CPU Cooler
-Corsair SSD
-Zalman Z11 HP Case w/ 5-120mm casefans and 2-80mm casefans

I have been having trouble with this computer build playing games since I built it. All of the drivers are up to date and the BIOS is up to date. I know that it can handle all of the games that I throw at it. Basically what happens is that I will be playing a game, FarCry 3 for example, and it will run fine on ultra settings for a little bit to a long time and randomly the screen will go black and the game program will become none-responsive. I ctrl+alt+del to the task manager and have to terminate the program. This problem is consistent and happens every time with varying amounts of time between when I start the game and when it freezes. I checked my processor temps and they were all below 38 degrees C right after the freeze. I went into the BIOS and changed the following settings hoping that it was a freak power issue or BIOS compatibility issue:

-Disabled Intel Speed Step Technology
-Disabled Turbo Mode
-Changed PCI-Express X16_3 Slot Bandwidth to X4 Moder
-Changed the PCI-Express X16_1 Speed to Gen2

Changing those settings didn't change anything in regard to my issue, and actually made my computer react a little erratically so I reset my BIOS to default.

I am at a loss for solutions here. I looked around and found that it may be a possibility that my PSU lacks the proper amperage in order to run these games at load; but I am skeptical because wouldn't that mean that my whole computer would crash and not just the game itself if I was too cranked into the amperage? I looked into it and used the eXtreme Outer Vision PSU Wattage and Amperage Calculator and for my system it recommended the following specs:

-803 Watt PSU
- +3.3V: 9.5A
- +5V: 14.8A
- +12V: 58.1A
-UPS

I have an Ultra 750 Watt PSU currently that's pretty old... probably new from 2007 and has been used heavily since then. Here are it's amperage specs:

- +3.3V: Max Load - 24A
- +5V: Max Load - 28A
- +12V1/2: 25A

Like I said, I am not sure if the PSU would be the problem since only the program stops responding and the entire computer doesn't crash. Perhaps the computer has a means of knowing which program is causing a load that the PSU can't handle and the computer makes the program crash? I don't know.

I am desperate to try and fix this problem. I invested a lot of money into this computer build and I can't even use it for what it was intended to be used for because of some kind of mystery problem. I checked around and it looks like some people with the same mobo have been having the same problem, so I think it might be a BIOS issue, but who is to know? It's obviously not a problem for all p8z77 owners.

I really appreciate any insight into the problem. Thanks!

Chris
a c 102 Ĉ ASUS
September 6, 2013 6:48:35 PM

Run OCCT GPU test and CPU tests for 60 minutes....note voltages and temps (can monitor BUS voltaGes w/ HW MONITOR)
m
0
l
a c 866 Ĉ ASUS
September 6, 2013 6:50:35 PM

Can you borrow a different PSU to try, sounds like it may be the problem, but hard to tell,
m
0
l
Related resources
September 7, 2013 5:59:35 AM

JackNaylorPE said:
Run OCCT GPU test and CPU tests for 60 minutes....note voltages and temps (can monitor BUS voltaGes w/ HW MONITOR)


I'm pretty savvy but not savvy enough yet to understand what most of that application is telling me.

I also don't know anyone who has a big enough PSU that I could borrow.

It's so tough because it doesn't even give me an error message.
m
0
l

Best solution

a c 102 Ĉ ASUS
September 7, 2013 8:40:46 AM

Well if you can pass 60 minutes in OCCT it's a perty sure bet that it's not a hardware issue.

Also..... the one thing that hurts OCs and hurts stability is varying voltages ..... CPUs / GPus have power control circuitry. What's happening when you have an OC'd CPU and it crashes is let's say ya voltage target is 1.25 .... well during varying loads, that voltage may vary say from 1.21 to 1.29

But oops..... at 1.21 it's not stable and crashes..... so ya have to boost it to 1.29....just so the low range voltage is 1.25

Higher end MoBos have sophisticated multi phase Power control VRMs which do this job better than the less expensive MoBos which is why the "affordable MoBos are just as good ass expensive MoBos" mantra is....well hooey. But as good as the MoBo power control is, it's job gets harder if the PSU voltage supply is all over the place.

I look for 1% voltage variation on PSUs (most jonnyguru 10.0 performance rated PSUs will fit this category) .... the ATX spec allows for 5% ..... I have seen several reports of late here on THG of CX series PSUs from corsair dropping below -5%.

Now when voltage drops 5% ..... amperage rises 5% ..... more amps more heat. That and .... well maybe ya getting 12.2 at low load and 11.35 at hi loads and your power control circuitry and MoBo, GPUs is scrambling all over the place trying to adjust and produce 12 volts when the load is jumping all over the place.

So....if ya running a loading utility, ya wanna see if where ya 3.3, 5 and 12 volt rails are and thereby determine if that's contributing to the problem.
Share
September 7, 2013 9:26:01 AM

JackNaylorPE said:
Well if you can pass 60 minutes in OCCT it's a perty sure bet that it's not a hardware issue.

Also..... the one thing that hurts OCs and hurts stability is varying voltages ..... CPUs / GPus have power control circuitry. What's happening when you have an OC'd CPU and it crashes is let's say ya voltage target is 1.25 .... well during varying loads, that voltage may vary say from 1.21 to 1.29

But oops..... at 1.21 it's not stable and crashes..... so ya have to boost it to 1.29....just so the low range voltage is 1.25

Higher end MoBos have sophisticated multi phase Power control VRMs which do this job better than the less expensive MoBos which is why the "affordable MoBos are just as good ass expensive MoBos" mantra is....well hooey. But as good as the MoBo power control is, it's job gets harder if the PSU voltage supply is all over the place.

I look for 1% voltage variation on PSUs (most jonnyguru 10.0 performance rated PSUs will fit this category) .... the ATX spec allows for 5% ..... I have seen several reports of late here on THG of CX series PSUs from corsair dropping below -5%.

Now when voltage drops 5% ..... amperage rises 5% ..... more amps more heat. That and .... well maybe ya getting 12.2 at low load and 11.35 at hi loads and your power control circuitry and MoBo, GPUs is scrambling all over the place trying to adjust and produce 12 volts when the load is jumping all over the place.

So....if ya running a loading utility, ya wanna see if where ya 3.3, 5 and 12 volt rails are and thereby determine if that's contributing to the problem.


I think I'm understanding you. So basically you're saying that if my PSU has poor amperage/voltage variation (which I am assuming by today's standards it probably does since it's from 2007) my components could be getting inconsistent or poor amp/volt combinations which in turn could be confusing them just enough to mess up the program I am running but not necessarily crash my entire rig?

I ordered a new PSU (OCZ ZX 850W) today which's 1000W counterpart from the same model line got a 9.5/10 rating at Johnny Guru anyways just to see if it changes anything.

I'll run that test, though, and see what I come up with.
m
0
l
a c 102 Ĉ ASUS
September 7, 2013 9:30:13 AM

I don't know if Id use the word confusing ..... Im thinking more like on of those tennis ball machines firing balls in different directions so fast ya just can't keep up.

m
0
l
September 7, 2013 10:21:15 AM

JackNaylorPE said:
I don't know if Id use the word confusing ..... Im thinking more like on of those tennis ball machines firing balls in different directions so fast ya just can't keep up.



You have been so incredibly informative in this thread. I really appreciate the time and effort you are taking to help me with this issue! Thanks.

I ran the test and I had an error approximately 8 minutes in. The error was: Core #0 Over Maximum Value. Value reached: 127 - Max value: 85

The value in question is the temperature of the core. I was looking through the different voltage data graphs to see if there was any correlations between those graphs and core #0's temperature graph and I found that at exactly the time of the temperature spike, my voltage AVCC (?) value went from 3.4 all the way down to zero for an instant. Here are the graphs:





I don't know what AVCC means, but it seems pretty clear cut to me now that my power supply is messing up somehow with the voltage/amp regulation. Do you also think that this is a valid assumption with these results?

Thanks again, you've been really great!


m
0
l
a c 102 Ĉ ASUS
September 7, 2013 10:27:26 AM

Well .... frankly I don't know how Intel protects its CPUs .... but maybe the safety circuitry cuts CPU voltage to make sure it doesn't get damaged.

What were the 3.3, 5 and 12 v rails doing at that point (HW Monitor)
m
0
l
September 7, 2013 11:50:58 AM

JackNaylorPE said:
Well .... frankly I don't know how Intel protects its CPUs .... but maybe the safety circuitry cuts CPU voltage to make sure it doesn't get damaged.

What were the 3.3, 5 and 12 v rails doing at that point (HW Monitor)


I'm no sure which graph on the readout will tell you that, so here is a link to an album containing all of the ones that it generated:

http://imageshack.us/g/839/c63y.png/

I appreciate you sticking with this topic.
m
0
l
September 10, 2013 6:31:57 PM

Update: I put the new PSU in and it did not change anything. Now I guess I can be pretty sure that it is either my motherboard or CPU that is defective. I am willing to bet that it is probably the motherboard because I am having some weird monitoring behavior from it. I will have to see about calling ASUS on my warranty.

Thanks for your help.
m
0
l
!