Freezing during games. Could SSD be the problem?

[ferrari91169]

I'm experiencing freezing every 5-10 minutes in most games I play, and I'm wondering if this could be cause by the SSD, or perhaps something else. It happens about every 5-10 minutes when playing a game, and the video will freeze, and the audio can still be heard, but will just keep repeating the same noise/words that were playing when it froze. It usually will freeze for about 2-6 seconds I'd say and it seems like it may get progressively worse as time of playing increases, although not always.

One thing I can say is when I'm playing Borderlands, it seems to ALWAYS happen when I teleport to a new area. Though it happens in other spots too, but it's almost positive it'll happen when I attempt to enter a new area.

Not sure what else it could be, and I'm kind of limited on items to test my different parts with. My computer will also freeze just when browsing, or on the desktop, and become completely unresponsive, in which I'll have to do a hard shut down.

Also, a few times, if I leave my PC on overnight, it will have shut itself down (or crashed), and when I start it up, it brings up the message about "Windows shut down unexpectedly, yadda yadda", and not sure what's causing that either.

Here are the rest of my specs:

Intel i7-3770k (OC'd to 4.4GHz)
16GB's 1600MHz RAM
DVD Drive/Blu-Ray Drive
750-Watt Corsair PSU
GTX 275 896MB
Gigabyte Z77X-UD5H Motherboard
128GB Crucial M4 SSD (Main OS/Some Games/Some Programs)
Various 320GB-1TB Drives for Storage

Can't think of anything else that would be important.

Before saying that the CPU overclock is unstable, I can say that I still had this problem before ever overclocking the CPU. Just thought I'd mention that.

Thanks for any help!

 

[ferrari91169]

Okay, I guess I actually SOLVED the problem with Borderlands. I moved it from one of my slave drives (HDD) to my SSD, and now it seems to run fine. No more freezing or anything. I guess I probably should've tried that first, but I do have this same problem with LoL, that IS stored on my SSD, but still has the problem. So I always figured it wasn't the fact that I was storing the game on a slave HDD, instead of my SSD, but I guess it was.

However, the system itself freezing, and the system "crashing" sometimes when left on all night, I still don't have any idea what's causing these problems, and I still could use help with those.

I have noticed the past two times I looked, that the last things that ran, in Event Viewer, were Adobe Flash Player Update Starting, and then Adobe Flash Player Update Stopping, within' a second of each other. And then after that I don't see any other events, so I'm assuming that's around when my computer shut off. Not sure why this would happen though, especially since it's been doing it across multiple installs of Windows/Adobe. So if for some reason Adobe was corrupted, I had reformatted and reinstalled Windows (and therefore Adobe), and I still have the problem.

But alas, twice now, that has been the last event running before my PC "unexpectedly" shuts down.

Anyway, thanks again for any help!

 

[cin19]

Maybe the temperature of your PC parts, like GPU and CPU ( because you OC to 4.4), using the software to check those temp.

 

[internetlad]

I'd update drivers. IIRC I had the same sort of issue a while back and it turned out that a Hard Disk driver was set up wrong. I regret that I can't recall the exact issue, it was years ago. IIRC It was something about being set up as IDE/SATA incorrectly but that might not be the case.

Random short term stuttering like that is an odd issue. I would say it's unlikely that the SSD would be the culprit unless A) It's experiencing a hardware issue and B) You're playing a game that dynamically loads level content from the drive (A LA WoW)

As far as any issues with random freezing/crashing/BSODS, boot into Memtest+ ( http://www.memtest.org/#downiso ) for a few hours and see what it spits out. When somebody says "Freezing" I think RAM.

You could also run a benchmark like BurnInTest and see if stressing the machine makes it worse, if so stress specific components until you see one that greatly increases the liklihood of failure.

That's where I'd start.

 

[DarkOutlaw]

I agree sounds like it might be an issue with your clock. Whats your voltage?

Could also be an issue with your windows install. How exactly did you install it? How many additional hard drives do you have?

 

[ferrari91169]

Maybe the temperature of your PC parts, like GPU and CPU ( because you OC to 4.4), using the software to check those temp.

The temperatures look fine for everything when I start up, and while playing games, etc. They never get too hot, from when I'm monitoring them. Also, the problem of the computer entirely freezing and needing to be hard-shutdown happens more often when I'm just idling, or browsing the internet, etc. So during those times, the temperatures should be more than okay.

I'd update drivers. IIRC I had the same sort of issue a while back and it turned out that a Hard Disk driver was set up wrong. I regret that I can't recall the exact issue, it was years ago. IIRC It was something about being set up as IDE/SATA incorrectly but that might not be the case.

Random short term stuttering like that is an odd issue. I would say it's unlikely that the SSD would be the culprit unless A) It's experiencing a hardware issue and B) You're playing a game that dynamically loads level content from the drive (A LA WoW)

As far as any issues with random freezing/crashing/BSODS, boot into Memtest+ ( http://www.memtest.org/#downiso ) for a few hours and see what it spits out. When somebody says "Freezing" I think RAM.

You could also run a benchmark like BurnInTest and see if stressing the machine makes it worse, if so stress specific components until you see one that greatly increases the liklihood of failure.

That's where I'd start.

I'll try running Memtest today. Should I run it one stick at a time, or just do all sticks at the same time? I have upgraded the firmware on my SSD thinking that could solve the problem, but it did nothing for me.

And I guess the short term stuttering is solved if I install the game directly to my SSD instead of the HDD. Although, as I said, I have noticed it while playing LoL as well, and LoL has always been on my SSD, so not really sure why moving Borderlands to my SSD fixed the short term stuttering in that game, but for some reason, LoL still has it on my SSD on occasion.

I agree sounds like it might be an issue with your clock. Whats your voltage?

Could also be an issue with your windows install. How exactly did you install it? How many additional hard drives do you have?

Well, I was having this issue before I overclocked my CPU, so that's why I don't think my OC is the problem. I never changed the voltage though, just ramped up the multiplier. But like I said, I had this exact problem before OC'ing my CPU, so I can't see why that would be the culprit.

As far as Windows...I installed it from a Windows disc, just the normal way. Stuck it in the drive, booted up, followed on-screen instructions, and just let it install, and that's it. I've also tried reformatting, and I borrowed my friends Windows disc, just in case my disc was corrupted for whatever reason, but still had the same problem using his disc.

As far a as SSD's/HDD's, I have four installed total. My main drive is an SSD (128GB), then I have a 160GB HDD (5400RPM), 320GB HDD (5400RPM), and a 1TB HDD (7200RPM). I started out with just the SSD and the 1TB for storage. But I upgraded both my netbook and laptop to SSD's, so I put the two drives from both of those in the computer for random storage.

 

[internetlad]

I'll try running Memtest today. Should I run it one stick at a time, or just do all sticks at the same time?

I'd say just do all the sticks at once. If it spits out errors, reseat all the sticks, dust out the slots and try again. If it persists try one by one, preferably in the original DIMM slots, then cross check the bad stick to a different DIMM slot. That would help narrow down if it's the slot or the stick itself, assuming it is the RAM at all.

It helps to know that you've tried with multiple installs, that would rule that out. To be honest we might be seeing two separate issues with the freezing and the stuttering. I assume LoL is installed through steam, have you tried verifying the game cache? Has the issue persisted in LoL through multiple installs? is LoL the only game that it's happening in? We might consider checking to see if there are any known performance issues with LoL doing specifically what we're seeing.

When you installed windows did you get the latest drivers off the disk or off the manufacturer's website? If you did them all off the disk, I might recommend updating to the latest from the site.


Past that I'd run a checkdisk ( Chkdsk /f at command prompt ) and if you're very through, a System File Check (SFC /Scannow) just to eliminate all the possibilities that there is something amiss with the files on the drive.

 

[ferrari91169]

I'd say just do all the sticks at once. If it spits out errors, reseat all the sticks, dust out the slots and try again. If it persists try one by one, preferably in the original DIMM slots, then cross check the bad stick to a different DIMM slot. That would help narrow down if it's the slot or the stick itself, assuming it is the RAM at all.

It helps to know that you've tried with multiple installs, that would rule that out. To be honest we might be seeing two separate issues with the freezing and the stuttering. I assume LoL is installed through steam, have you tried verifying the game cache? Has the issue persisted in LoL through multiple installs? is LoL the only game that it's happening in? We might consider checking to see if there are any known performance issues with LoL doing specifically what we're seeing.

When you installed windows did you get the latest drivers off the disk or off the manufacturer's website? If you did them all off the disk, I might recommend updating to the latest from the site.


Past that I'd run a checkdisk ( Chkdsk /f at command prompt ) and if you're very through, a System File Check (SFC /Scannow) just to eliminate all the possibilities that there is something amiss with the files on the drive.

LoL is actually not installed through Steam, it's not available on Steam. Haven't seen any one complaining about similar problems in LoL when looking online, but that doesn't mean it doesn't exist for other people.
And yes, the issue has persisted in LoL through multiple installs.

When I installed everything, I did use Windows Update, but didn't do much besides that. I did get the updated driver for my video card, and most recent firmware for my SSD (to see if that solved the problem, which it didn't).

I'm about to run a memtest now, and I'll try and get back with the results ASAP.

 

[ferrari91169]

Okay. So I'm 70% done with the first pass at the moment (50 minutes in), and haven't gotten any errors so far.

Using Memtest86+ v4.20

I wanted to ask though. Is it important to do multiple passes, or can you be pretty sure after one pass that your RAM is good? I'm guessing the first pass is going to take around 2 hours minimum, and not sure how long I want, or need, to have it running.

I'd also like to mention, it seemed to be doing even worse today with the random shutdowns. Usually it only does random shutdowns when I leave it on overnight, and I usually experience them at most three times a week I'd say. But so far to day it's shut itself down (or crashed) five times so far. Once it shut itself down about 15 seconds after I turned it back on and it was at the Windows loading screen.

 

[DarkOutlaw]

When you did the install did you only have the SSD connected? Or did you install with both drives connected?

 

[ferrari91169]

When you did the install did you only have the SSD connected? Or did you install with both drives connected?

The last time that I installed Windows, I had all the drives connected IIRC. The time before that though, I can't really recall. I may have only had the SSD connected that time and then connected the others after Windows had installed.

 

[internetlad]

random shutdowns could be PSU. Badly regulated voltages could also cause a myriad of issues. It's less likely with a name brand PSU, But I wouldn't trust anything 100% at this point. Any chance you can test the voltages? PSU testers aren't that expensive, but it would take a couple days to get it shipped. Alternatively, any PC repair shop worth it's salt should be able to test the PSU for a nominal fee.

One pass of Memtest should be fine. If you don't see any issues arise it should indicate that the RAM is okay. I'd still reseat it just in case.

EDIT: Thinking about it, Random shutdowns are a pretty telltale sign something big is up hardware wise. There aren't many components that will just cause the entire unit to take a dump. Most components will generate a BSOD when they go, or cause a failure to perform in the OS but not actually shut the whole machine down. You'd probably recieve a "thermal event" notification if it was overheating.

In my opinion, I'd look at the PSU or, worst case scenario, the Mobo at this point. For simplicity's sake, i'd test/swap the PSU first. If that fails, there aren't many options beside the board.

For troubleshooting's sake, you could try unplugging everything but the SSD and running it like that. No optical, no USBs, start up the board by shorting the 2 power pins with a flathead and run it with just the add in video if (it doesn't have onboard) and the boot drive and see if you still have the issue. Intermittent issues like this are always a pain in the ass.

Hope that helps, best of luck.

 

[ferrari91169]

UPDATE:

Just ran Memtest 86+ v4.20 for about 8 hours/8 Passes, and came up with 0 errors. Not sure if I should run it longer than this, as I know it can sometimes take 48 hours+ to show errors, but those are the results thus far.

random shutdowns could be PSU. Badly regulated voltages could also cause a myriad of issues. It's less likely with a name brand PSU, But I wouldn't trust anything 100% at this point. Any chance you can test the voltages? PSU testers aren't that expensive, but it would take a couple days to get it shipped. Alternatively, any PC repair shop worth it's salt should be able to test the PSU for a nominal fee.

One pass of Memtest should be fine. If you don't see any issues arise it should indicate that the RAM is okay. I'd still reseat it just in case.

EDIT: Thinking about it, Random shutdowns are a pretty telltale sign something big is up hardware wise. There aren't many components that will just cause the entire unit to take a dump. Most components will generate a BSOD when they go, or cause a failure to perform in the OS but not actually shut the whole machine down. You'd probably recieve a "thermal event" notification if it was overheating.

In my opinion, I'd look at the PSU or, worst case scenario, the Mobo at this point. For simplicity's sake, i'd test/swap the PSU first. If that fails, there aren't many options beside the board.

For troubleshooting's sake, you could try unplugging everything but the SSD and running it like that. No optical, no USBs, start up the board by shorting the 2 power pins with a flathead and run it with just the add in video if (it doesn't have onboard) and the boot drive and see if you still have the issue. Intermittent issues like this are always a pain in the ass.

Hope that helps, best of luck.

I've actually already switched the PSU out before, as I did have an Antec 600-Watt, and I switched it out for the Corsair 750-Watt within' the past month. I was experiencing the issue with both PSU's, so I'd assume it wasn't a PSU problem...unless I'm really unlucky and got two PSU's with the exact same problem. I do not have a way to test the voltages though.

I've now run the RAM in Memtest for about 8 Passes/8 Hours and received 0 hours.

As far as the BSOD's, etc., that was my thought exactly. I've never had a hardware failure that didn't supply a BSOD before...at least that I remember. And I've never had an overheating event that didn't somehow notify me about the computer overheating. So it was quite weird when I started having this problem.

No good way to test my Mobo I suppose besides switching it out, right? Or my CPU?

I will be able to test my GPU soon, as I just ordered my new card. The one thing I'm wondering though. If I put a new GPU in, and someone else is causing the system to shutdown...what are the chances it could damage my new GPU?

 

[cin19]

When I look for something else I got this from other forums. Try this: 1)set pck clock to 101 instead of auto
2) Turn off c6/c3 in the cpu advanced features.

Hope this will help.

 

[internetlad]

Here's something you can try really quick. Mash the F8 button at startup and choose "disable automatic restart on system failure" (or something close to it)

I had a machine that was experiencing mysterious rebooting, and it wasn't showing the BSODS, enable this and it will display them until you turn the machine off.

And no, in my experience the only thing you can do is breadboard the mobo (IE remove it from the case to compensate for any shorts, remove all but the necessities and lay it flat on an antistatic surface, cutting open the Anti static bag it shipped in works well.) and see if the problem persists.

Past that, we have to assume from process of elimination that your mobo is moboned.

 

[internetlad]

I can't remember what it was called, but I didn't find a CPU testing utility that i used for a couple months a while back. It didn't work well enough for me to keep it in my arsenal though.

CPUtest? something like that? A quick google search is showing one repeatedly, CPU Stability test 7.0. This isn't something I've used before.

Have you tried running BurnInTest? It's a very versatile tool. I've gone so far as to purchase the full software.

 

[ferrari91169]

Here's something you can try really quick. Mash the F8 button at startup and choose "disable automatic restart on system failure" (or something close to it)

I had a machine that was experiencing mysterious rebooting, and it wasn't showing the BSODS, enable this and it will display them until you turn the machine off.

And no, in my experience the only thing you can do is breadboard the mobo (IE remove it from the case to compensate for any shorts, remove all but the necessities and lay it flat on an antistatic surface, cutting open the Anti static bag it shipped in works well.) and see if the problem persists.

Past that, we have to assume from process of elimination that your mobo is moboned.

I just started up Windows with the option you suggested, so I'll see what happens. Not sure if anything will actually display though. From the times this has happened before, Windows never saves a BSOD Log, which I have it set to, so I guess it's not generating one. And also, from seeing it happen twice now, first hand. All that happens is the computer is showing on my monitor one second...like the desktop...then one millisecond later it just goes black and the computer goes from running, to a complete shutdown. There's no messages, nothing loading, nothing happening...no freezing and then three seconds later it shuts down. I can literally be moving the cursor around, and it just goes black...as if the powers gone out.

But like I said, I did what you suggested, and I'll see if anything of significance is on the screen next time it happens.

When I look for something else I got this from other forums. Try this: 1)set pck clock to 101 instead of auto
2) Turn off c6/c3 in the cpu advanced features.

Hope this will help.

Not exactly sure what the PCK clock, is, or C6/C3 is, but I'll look in bios next time I restart. Did you mean PCB Clock? I'm not even sure that's a thing, but I'm just throwing something out there. In any case, I'll look.

I can't remember what it was called, but I didn't find a CPU testing utility that i used for a couple months a while back. It didn't work well enough for me to keep it in my arsenal though.

CPUtest? something like that? A quick google search is showing one repeatedly, CPU Stability test 7.0. This isn't something I've used before.

Have you tried running BurnInTest? It's a very versatile tool. I've gone so far as to purchase the full software.

I haven't tried BurnInTest yet. I suppose I could try that later...although seeing as a lot of the times the freezes/shutdowns happen when I'm not doing anything CPU intensive, I'm not sure BurnInTest would produce any help. BurnInTest is used to check stability of your CPU under an intensive load, IIRC.

But next time I get a chance, I'll look into that.



UPDATE:

Would like to mention that I downclocked my CPU back to stock, as well as my RAM. My RAM wasn't really overclocked, I just had to go in to my bios and choose "Profile 1" to get it to run at 1600MHz, instead of 1333MHz.

But right now they're all back at default, so maybe that will help.

 

[cin19]

IF you use GIGABYTE 3D BIOS, go to Frequency to change CPU/PCIe Base Clock to 100.1MHz instead of auto (100MHz).

http://images.anandtech.com/doci/6108/Gigabyte%20Z77X-UD5H%20BIOS%2002%20-%203D%20CPU.png

 

[internetlad]

he means the CPU clock.

 

[ferrari91169]

UPDATE:

Not that this means anything really, but I had my computer on overnight, downloading a game from Steam, and it hasn't shut down yet. Again, this doesn't really mean anything, as it's never really been an exact measurement of when it shuts down. Some times it has gone 24 hours+ without being shutdown, and sometimes it's gone multiple days without any problems.

Anyway, I've downclocked my CPU back to stock, and I've set my RAM at 1333MHz, instead of 1600MHz.

I've also set my CPU/PCIe Base Clock to 100.1MHz, as someone suggested. Whether any of these fixed the problem, or whether it's just wishful thinking...only time will tell.

Thanks thus far for all the help though. I'll update with any new information.

 

[internetlad]

Would you even notice a 266 mhz clock increase on ram? What would that be like 1 fps? I've always regarded modern Ram clocks/OC'ing as greatly overappreciated.

 

[ferrari91169]

Would you even notice a 266 mhz clock increase on ram? What would that be like 1 fps? I've always regarded modern Ram clocks/OC'ing as greatly overappreciated.

I always felt I could notice a slight difference between 1333MHz and 1600MHz. Not to mention WEI gives my RAM a 7.9 @ 1600, but only a 7.7 @ 1333. In any case, I always do notice a very slight difference, nothing game changing. Anything after 1600MHz though I can't notice any difference whatsoever, but that could just be me.

 

[ferrari91169]

Just another quick update. I have successfully left the PC on 3 or 4 nights now, with no problems whatsoever. Its exactly like I left it before going to bed. Not sure what fixed the problem (or if it's fixed), but for now it seems to be good.

Thanks for all the help!

 

[internetlad]

Glad you got it solved.

 

[ferrari91169]

Best answer selected by ferrari91169.