Ever since getting my new computer, it's had this problem: Around every month or so, when playing games, it will just randomly lock-up. The freeze is purely process-based, as far as I can tell. Despite the game freezing, the OS itself continues to run and I can just close the game out. Having had about 10 months of experience with this, I know the symptoms of it very well: The game's image freezes, but sounds that were playing at the time continue to play through or loop properly (no stuttering or such). Music tracks will play through, VOs play through, ambient noise continues to loop, etc. I've extensively recorded instances of this occurring in the past months through EVGA precision. FPS (predictably) drops to 0, and pretty much everything else drops off slowly after the freeze (gpu temp, clock speeds, etc). The only things that don't change are GPU memory usage and Memory Clock. I'd include links to spreadsheets of what this looks like to EVGA precision, but I cannot post links yet. There's seemingly no pattern in when or where the freeze happens. No temperatures seem too high (Maximum Temperature on GPU I've seen is 79C, and it's certainly not when it's that hot when the freeze happens - CPU wise, hottest I've seen is around 50-60C), and it doesn't even seem tied to how long I've been playing: I've had it happen anywhere between 10-20 minutes into playing to 5 hours into playing. The closest thing I can figure to be related is that it has something to do with loading - which lead me to test my RAM and Hard Drive several times, only to discover absolutely nothing wrong with them. I've seen it occur in a wide number of games: Fallout New Vegas, Skyrim, Saints Row IV, Champions Online, and Team Fortress 2 technically (for some reason, whenever it seems like it might lockup, it ends up crashing to desktop instead). All of these games are run through Steam. From what I can tell, while the issue is not hardware related, it also seemed to correlate a bit with GPU overclock: While I was using the system with a superclocked PNY card (which had another overclock over that to start), it would freeze after about 4 hours of playing between two games. After underclocking (first to its factory settings, and then attempting to get back to normal), it'd happen around maybe twice a week. After I managed to get the PNY card RMAed and got a replacement, it now happens extremely sporadically. Related to that: I have tested the computer using Valley and Heaven GPU benchmarks (~8 hours of running), Prime95, Memtest86, Windows Memory testing utility, and Seagate harddrive tools. All tests were completed without issue. To drive the point home about this being pretty much entirely agnostic from hardware load, there was one instance where a freeze occurred in Champions Online in the middle of a battle. In response to this, I proceeded to open up the game again in Steam, log on, and successfully finish the battle. I had two copies of the game process open: one frozen, the other running completely fine. One was a running game, and the other was literally a frozen ambient sound-playing screenshot of the same game from about a minute ago. I closed it out and continued playing with absolutely no hitches for 4 hours. :bugeye: I suspected it might be due to video drivers. I went through various different NVidia video drivers, usually with very little extra software options (just driver and physx essentially), but regardless of the driver it seems to always happen eventually. Other people have certainly voiced issues about NVidia Drivers recently having micro-stutter. Which is not the issue I'm having; I'm having eternal game-stopping stutter. I've gone to various other sites and forums to try and figure out a solution to my issue, with advice alternating between utter silence and 'instantly reinstall your operating system'. While I'd normally be open to the latter, I am pretty sure the solution to the problem is something simple to do with bloatware, and quite frankly I don't want to reinstall my entire system over something that most people would probably brush off as an extremely minor problem. I'm certain it's got something to do with finding the right driver, system events, or some weird incorrect installation of Steam. Any help or reply will be greatly appreciated. EDIT: Since I now have enough posts for it, here's what EVGA Precision recorded for my GPU when some freezes happened: Fallout New Vegas: Champions Online: The memory usage change was likely due to Google Chrome running in the background - Chrome I know for a fact takes up some VRAM when it runs with Acceleration.
Another card to test on? Do you have any graphic card to test on? Did you try seating the graphic card in another PCI-e x8/16 slot? Let me know. Did you check and see if you're getting enough power via rail amp?
Hey, thanks for the reply! All I have for alternatives is my integrated Intel HD4600 or whatever. I actually ran with that for a couple of days (which was a huge pain due to horrible game lag) and didn't get any issues like this one. I also have an ATI Radeon 4850 from a 5-year old computer, but It's not Dx-11 Compatible, and probably won't run on Win8. My PSU should be more than powerful enough to run it. I'm not overclocking or anything, and when running an intense game (that keeps it at full boost most of the time) it's outputting 1.2 Volts according to Precision. Not sure if that's good or not, but that seems like the appropriate max-power for such a rail, and my games usually run fine with that. Seating the card into another PCI slot, though. Haven't tried that. Will it still read the card as my first one if I put it into the second slot, though?
So, I just had it happen again. Incidentally exactly one month after the last one. (The last one was on October 1st, this one's on November 1st, how about that). A few things to note that changed between then and now: I had read some factoids on someone else with a similar problem with their NVidia GPU; similar sporadic freezes with sound continuing, but ending up in a TDR. Suggestions within that thread advised the user to try underclocking, or otherwise reducing stress. Because of the improvements I'd seen between this GPU and my previous one (which was superclocked), This lead me to try underclocking the GPU to disabling boost (Core Clock Speed never goes over 1045MHz). This initially ran fine for a while - similar as before, and in fact with basically no performance loss ingame. However there were a couple of times hiccups seemed to occur where both the visuals and sound on my computer would freeze momentarily - this was all system sound, by the way. Even things playing on WMPlayer in the background were affected. Another thing I noted while this was going on was that voltage was at 1.2V when at full default clock. Prior to that, whenever Clock Speed got up to 1045 MHZ the voltage was closer to 1.111 or so. As for the exact freeze today, it occurred maybe around an hour and 30-40 minutes after game startup, and happened either right after I opened an ingame player-search window, or after I hovered over the side of the window for a split-second, since the mouse cursor was frozen on a horizontal-resize graphic. For some strange reason a lot of these freezes I get seem related to UI transitions or something. That aside, we can safely cross out the possibility that it's caused by too little voltage. I've also been running HWInfo alongside EVGA Precision to try and get an idea of what the rest of the computer is doing. The biggest culprit I could figure, besides the GPU, is the hard drive, which didn't seem to be doing anything bizarre when the freeze occurred. The really periodic nature of this freeze is really beginning to make me think that it has to do something with some really intrusive/invisible program that's trying to update once a month, but I can't think of anything that really could be doing that, and the fact that the computer's practically been doing this since I got it leads me to think otherwise. If it helps at all, I can attempt to see what a hijackthis log reveals?
Bump again, sorry. It's happened yet again now. Since then there's been only a few changes: Drivers were rolled back from 337.88 to 331.93 - which seem even less hiccupy than the former ones I used, which is good, but the eternal freeze still showed up of course. Despite this the full sound/visual hitch still occurred - until I turned off hwmonitor's tracking. Since HWMonitor wasn't picking up anything suspicious in the logs. Most likely, the act of copying down like 40 columns of data every 2 seconds was a bit too much for the CPU to handle while also playing a game. That said, in times I was running HWMonitor while also running the game, I found no correlations between slowdown and sudden spikes in disc usage, so that's likely not an issue at all. That all said, I can't say I found anything to conflate the freeze to anything to do with hardware, since I wasn't tracking it at the time... Stupid me, trying to enjoy games. I also had updated my mobo BIOS and Ethernet drivers between then and now. I doubt that has anything to do with the issue, but at least it crosses out those software elements as culprits, I think. Despite the game locking up, the game process does not show up as 'not responding' in the task manager. Though, I'm almost certain this varies from game to game. At least in Champion's Online's case, the process doesn't think it's frozen. One thing I have noticed, though, that is likely unrelated - when the GPU gets hot, it seems like the front of my tower makes a clicking noise. Seems to usually come from the top of the tower (where my DvD drive is), though also sometimes lower down (more towards my HDD). Doesn't always make this sound, and it's only when I play games that get my GPU going hot. (and sometimes a while after quitting said games). I'm 99% sure that it's just how the tower is built, as naturally just pushing down a bit on the sides and front of the tower on those locations produces a bit of clicking. The tower model is an Enermax Fulmo mid-tower. Still am thinking that some odd software component is at play here - or just the GPU drivers themselves, in which case I'll somehow need to await completely flawless drivers for the 770 to come out or something like that.
I have one suggestion that may or may not help in any way with your issue. I had a similar problem years ago when I first built this PC. What was causing it for me was a conflict between Realtek and Nvidia's HD audio on the video card. The suggestion is to either disable your onboard sound card in BIOS if you're getting your audio from HDMI / Displayport from the video card, OR to disable the High Definition Audio Controller for the video card through device manager if you use your motherboard's onboard sound card. If it helps great, if not then at least it rules out a conflict between these two things.
Considering that I'm using my monitor through an HDMI cable, the fact that it shows up as an audio output device, and that those both have generally been quite consistent in usage since I first started using the computer, I think you might be onto something. I'll try by disabling the monitor as an audio playback device (since it clearly doesn't have speakers) and see what happens.
Well, got not 1, but 2 freezes in rapid succession tonight right now. It was really weird, because not only was this the first time I've had them happen in such a time lapse, but it was also while doing roughly the same thing in the same game. Changes between then and now: Disabled the HDMI monitor as a playback device (I mean, it doesn't even have speakers on it). Crosses that out as a potential culprit. At the very least, I hope that makes my weird issue of NPC speech in Skyrim randomly cutting off when I walk away from them not happen. Or less common. I believe I updated Razer Synapse a while back. It's mostly been running offline mode since I installed it. Doubt it's a huge factor, since I believe the very first freeze I got occurred before I attached and used Razer keyboard and mouse. However, the freezes I got (and have been getting) do seem slightly related to mouse input... Uninstalled Intel Rapid Storage Technology drivers, which were installed on my computer by the assembler. From what I researched, they're for RAID array system configurations. And my computer only has one hard drive. I haven't noticed any performance reduction since uninstalling, and had suspected that it might've been causing the freezes, since it has been one of the only constants since I got the system, but that's clearly not the case. Changed my computer-using habits a bit, as a test. Instead of shutting down/restarting my computer each day, I decided to try just keeping it on constantly and putting it to sleep at the end of each day. Over the past 2 months, with about only 3 or 4 boots, absolutely no freezing issues occurred. I just did one full shutdown yesterday (since the computer decided to wake itself up for some reason), and the first freeze in weeks occurred after I booted up from that. What's useful about this is that it at least proves that the number of games played in a boot doesn't seem to have any negative effect on performance. Though I went 2 months without anything happening, I did notice some things happening when I started playing Dark Souls: Sometimes, by random, the game would experience a full temporary lockup, not unlike the symptoms I've been describing. However, after a few seconds, the game starts up normally again. From what I gathered from that scenario, I guessed that this was due to me running the game as a Steamworks Beta, and that it still has some kinks with connectivity/invasions/messages appearing correctly to iron out. But I guess that would narrow the issue down to something related to either mouse/keyboard (since they are both unusual Razer items, after all) or my ethernet drivers. Will attempt to look into that; my guess is on the latter, since 2 rapid freezes occurring near the same time, doing a similar activity (attempting to open a trade window, and pressing an accept button on aforementioned trade window), seem to make that the most likely possibility.
Have you tried a fresh OS installation? Also i would make sure you have the latest bios and disable everything you dont use, ie 3rd party storage controller, usb 3, IGP, etc.. Install the latest drivers for RST, chipset(should be version 10.x) and nic http://station-drivers.com/index.php/downloads/Drivers/Intel/
Did you disable your secondary sound device in device manager like this? Example: That specifically is what stopped the freezing / driver crashing for me. If you did then carry on.
Fresh OS installation's something I've been consciously trying to avoid, since to me it seems like a disproportionately drastic solution to something I feel most normal people would dismiss as something insignificant. It's also been happening to me since day 1, so what seems likely to me in that scenario is just getting everything set up again and all my steam games downloaded over 2 or 3 days, just to run into the same problem again. Of course, if nothing else works, that'll be what I try next. I've been keeping up to date on BIOS by checking the ASUS site now and again. I'll see what I can find on my chipset and net-interface-controller, but I don't feel like RST really was doing anything for me. EDIT: So for the NIC, the category I'd look under is Network Connections Software, right? Most of the stuff I've gotten from ASUS is for a Realtek LAN contrller (which, knowing my luck, is probably something similar but different). I did go into the device manager to disable "High Definition Audio Device" under "Sound, Video, and Game controllers", but I totally missed over that option under the System Devices category. Much like you, it looks like I have two of those. Should I try disabling both of them or just one? In any case, hope that's the solution to my problem! EDIT: Ok, found my answer to that question. Disable the one that doesn't cause my system's audio to stop working.
