I've just finished building my new PC build i7-6700k asus z170-A, and windows 7. And I have a problem. Every so often the computer freezes. Completely locks up and requires a hard reset. As it freezes, whichever audio device is outputting at the time let's out what I can only describe as a horrible death screech. Originally googling suggested it might be the realtek audio driver, so I removed that but its still happening. Has anyone experienced something like this?
Windows 7? Why senpai?
>>51280848
Complete spec would be helpful.
>>51280971
This
>>51280971
Seconding this. Every single thing plugged in to.your PC when it freezes, pls.
>>51280848
Happens all the time when my shit freezes when running my unstable GPU overclock.
Yup, post full spec. How many modules of RAM are you using? If more than 1, have you tried removing them to eliminate a memory issue?
>>51280848
voltage issue
>>51280971
I7-6700k
Asus Z170-A
ASUS STRIX GTX970
Corsiar vengeance LPX DDR4 @2133MHZ 16GB
Corsair RM850
OCZ vertex OS drive, Sandisk SSD, and 2 mechanicals.
>>51280990
Ok, on top of the parts there is a corsair h55 cooler with fan. 2 140 top fans 2 120 front fans (that don't seem to spin properly. Now starting to wonder if that's related) plugged in the back I have a corsair 1500 v2 headset, the logitech usb receiver cable and the USB charge cable. 2 monitors 1 HDMI one display port. Network cable and 3.5mm aux jack to my speakers.
>>51281164
First thing I'd dois reinstall your network drivers. It may sound weird but trust me.
>>51281127
>Corsiar vengeance LPX DDR4 @2133MHZ 16GB
There's your problem, most likely.
I have the same RAM and it doesn't fucking work at its rated speeds, you have to underclock it. Since then my Skylake system has been working 100% stable.
>>51281182
will try that. meanwhile heres hwmonitors voltages
>>51280848
Had this problem a month ago on w7 on my laptop, exactly what you said
Turned out my pc was trying to eject my gpu, had this error message one time before he crashed.
So I deactivated my gpu and reactivated it in Windows and everything worked after that, so try it.
Maybe that was in an update meant to intentionally fuck up w7 to make people switch to w10.
>>51281230
you might be onto something, to be honest though im used to older keyboard navigated straight forward bios layout. my new motherboards bios looks like something out of a fucking spaceship. not entirely sure how to navigate it let alone underclock my RAM.
>>51281182
done. unfortunately no way to test it until it happens again.
my RAM from speccy as well as a possible HDD issue.
>>51281444
forgot image
also i keep seeing this in the event log around the time of the crash "Driver ACPI returned invalid ID for a child device (5)."
>>51281481
>>51281250
Try what I said, it costs nothing.
I had this and it was bad ram
>>51281553
i would if i had any idea what you meant by deactivate the GPU. do you mean disable it from the device manager?
>>51280848
I had that exact same issue a month ago. After testing everything over and over.... I finally had a hunch that it was the mobo.
Brand new MSI mobo too... Fuck MSI.
Got rid of it and brought a gigabyte mobo. Everything else was the same as before. Now it works perfectly without errors.
>>51280848
i think it's to do with the motherboard as mine does the same and i thought it was down to using single channel memory.
>>51281637
It's not msi.
maybe im on to something here?
>>51280848
op definitely check memory with diagnostics thing in windows
>>51282040
ran windows memory diagnostics tool, found no issues as far as im aware. dont know where the logs get put though
>>51281230
That makes no sense at all for the issues that i had recently with linux distros.
I have almost the exact build as OP, only 2400MHz 16GB RAM sticks.
The whole system freezes like OP's problem under linux, no logs and no response from the machine whatsoever.
However it works fine under Windows 8.1....
>>51280848
That regularly happens with unstable CPU OCs for me, so I'd say it's likely you have some CPU/RAM related instability. First make sure everything is running at proper stock settings and isn't overheating, then try to stress test to see if you can cause it to happen. Make sure you run your RAM at stock settings supported by your CPU, even if your memory is actually rated higher.
disabling speedstep solved a very similar issue for me
>>51282217
i've got a 6600k and Z170-A and i get the same issue as OP with both the memory and CPU at stock settings, and the only thing for me which makes it happen is gaming.
>>51282063
>>51282213
is it some kind of driver issue maybe?
you need to test everything once systematically till the freezes stop and I believe there is no other way.
>>51282213
Linux was known to have some early lockup problems with Speedstep enabled, perhaps you suffered from that. A microcode update should fix it.
>>51280848
Take out the GPU and test. Then test with only one of the rams, until you have tested them all. At least this way you can single out the problem a bit.
>>51282254
It means your PC is unstable at stock settings, for some reason. Stock settings aren't infallible, both you and OP can probably fuck around with voltages and/or RAM settings and get a stable system. I had something similar happen with Haslel, turned out to be due to RAM settings.
>>51282260
I tried to turn off everything energy saving related options in BIOS, and even then i had the same issue.
>>51282263
I don't know, if i disabled speedstep in bios then that shouldn't really matter?
I could try to run the memory underclocked, but if it works on Windows that what is the point of dicking around with undervolting just to get linux running?
In the end there is something buggy with Skylake and linux.
>>51282323
i'm not sure about that as stock voltages go between 0.821 idle and 1.314, i'm not sure about it as it's stable at 4.2ghz with 1.25, it very well could be that skylake is really picky with memory or it could be related to the iGPU and shared memory, when i get back try i'll clear cmos and play with memory settings.
>>51282330
>In the end there is something buggy with Skylake and linux.
Not necessarily, though I don't know if it is or is not the case since Skylake is still quite new. Not all workloads are the same, it's completely possible to crash in certain scenarios and appear stable in others.
>>51282448
Could very well be due to RAM and not actual CPU settings. Try fucking around with memory & memory controller settings. Drop memory clock, slightly relax timings, run 2T command rate if you're on 1T, slightly bump all related voltages, see if it helps. If it does, figure out what exactly is making you crash by resetting settings individually.
I actually highly doubt it's related to core CPU clock, it's probably something related to memory.
>>51280848
I had that issue when I did a fresh reinstall of windows 8.1. I think the problem was that my hard drives were spinning up and down and kept causing the PC to lock up and sometimes needed a reboot. I changed it to never turn off or power down w/e and it stopped.
>>51282511
also when i get round to it i'll pick up a 2nd stick of ram as my issue could be coming from having single channel.
Weird OP, I tried asking for help here on /g/ a few years ago with my last pc, no one helped ayy lmao I guess.
I've had this weird lock-up shit happen on EVERY PC I've ever owned, ( pre-builds / my own custom made and such. ) This current pc is a lenovo pre-build. i5-3330 / 8GB / GTX 660ti / 600 watt corsair PSU. When that issue you descibred started to occur here. I found out it was Avast. I reinstalled it stopped.
Then it occured again and after much more effort, doing everything everyone above listed as work arounds. I found out it was the windows indexer causing the issue. I disabled it and have never had the issue occur again.
I'm planning to buy a new prebuild this week with windows 10, which I'm not looking forward to w10... As 8.1 never gave me any issues and was by far the best I'd used. All well, anywho upon getting that I'm disabling that fucking indexer hopefully... For once. This issue won't occur for me. Would be nice.
>>51282563
>my issue could be coming from having single channel.
No way in hell
>>51282239
speedstep undervolts, so it might be causing RAM instability
