PC shuts off when rendering video

Doogles · 28 Oct 2016 at 19:28

Combat squirrel said:
Ok this is weird. All the memory sticks work fine in duel channel mode, 3 or 4 sticks throw errors, what?

It could be the motherboard or it could be the IMC on the chip, have you tried lowering the settings to high latency and low frequency to see if it helps?

Combat squirrel · 28 Oct 2016 at 19:38

Doogles said:
It could be the motherboard or it could be the IMC on the chip, have you tried lowering the settings to high latency and low frequency to see if it helps?

Yeah iv lowered it to 1600 10-12-10-28, ram can do 9-11-9-27

Same thing, think the CPU memory controller is dying or is it the mainboard or ?

Doogles · 28 Oct 2016 at 19:46

Combat squirrel said:
Yeah iv lowered it to 1600 10-12-10-28, ram can do 9-11-9-27

Same thing, think the CPU memory controller is dying or is it the mainboard or ?

I have no idea and I think the only way to be sure is to change the CPU to test, do you have a spare PSU by any chance? I know it is a long shot but it could be possible that it is the culprit. One other thing have you taken off the CPU overclock, it can make the RAM unstable.

Combat squirrel · 28 Oct 2016 at 20:10

yeah its odd, running defaults duel channel now happy as larry, trying different slots for duel channel see if its a particular slot thats failed

Rroff · 28 Oct 2016 at 20:16

9 times out of 10 if the system is hard locking or rebooting when rendering video its due to not enough voltage either for the IMC or core - you can run Prime95 or something like that indefinitely and not necessarily cover instruction combinations that you will do encoding video. Next suspect being the PSU unable to provide the current required stably once warmed up.

Normally though no guarantee if it was faulty memory its more likely you'd just have the application crash or some other file corruption rather than taking out the whole system (in the context of media encoding).

EDIT: What voltage were you running the IMC/VTT (sometimes called system agent or VCCSA) - not sure what the sensible range is with the 4930K but the previous generation some people would degrade the CPU with more than 1.2v especially with frequent encoding, etc. type tasks.

Combat squirrel · 28 Oct 2016 at 20:19

Yeah it seems stable in duel channel for now, ill overclock it back but with duel channel and see what happens - if thats stable, we know something to do with 4 channel is afoot

Combat squirrel · 28 Oct 2016 at 21:00

Sa was like 1.1v, it's back to 4.6ghz now running just fine no problems with 16gb in duel channel mode.... Annoying loosin half my ram but works fine so far, testing the removed ram in another system

Combat squirrel · 29 Oct 2016 at 13:28

Stage 2: Running the memory with memtest in another machine for a few days to see if it throws errors - ill swap out the ram 2 at a time, if nothing comes back as a problem, at least I know the ram is ok.

I suppose then ill be on the hunt for a cheap 2nd hand x79 mobo as my next thing to test - don't suppose anyone has a x79 laying around I could rent ? haha

Doogles · 29 Oct 2016 at 14:23

Combat squirrel said:
Stage 2: Running the memory with memtest in another machine for a few days to see if it throws errors - ill swap out the ram 2 at a time, if nothing comes back as a problem, at least I know the ram is ok.

I suppose then ill be on the hunt for a cheap 2nd hand x79 mobo as my next thing to test - don't suppose anyone has a x79 laying around I could rent ? haha

Funny as yesterday I had a freeze on desktop and all I was doing is running my backup software and I had one long beep indicting that I have bad memory too, not soon after I was posting to you about you're issues, I think you are contagious! I'll ignore it myself for now but when I stop using my PC I'll run memtest myself for 12 hours.

Combat squirrel · 29 Oct 2016 at 15:03

Doogles said:
Funny as yesterday I had a freeze on desktop and all I was doing is running my backup software and I had one long beep indicting that I have bad memory too, not soon after I was posting to you about you're issues, I think you are contagious! I'll ignore it myself for now but when I stop using my PC I'll run memtest myself for 12 hours.

Strange isnt it !! finally wearing out the PC :eek:

theaterix · 29 Oct 2016 at 15:41

Had a shutdown issue a while ago, just cut off, I have a corsair CX750M and so I set the 5820k to stock aswell as ram and bam issue gone... May of been windows as it did not go instantly but it happened a lot less frequent. Give it a bash

Combat squirrel · 29 Oct 2016 at 19:35

theaterix said:
Had a shutdown issue a while ago, just cut off, I have a corsair CX750M and so I set the 5820k to stock aswell as ram and bam issue gone... May of been windows as it did not go instantly but it happened a lot less frequent. Give it a bash

Yeah have reset everything, still happens, either the board or CPU - more likely board - will shop about for one to test.

Doogles · 29 Oct 2016 at 20:46

Combat squirrel said:
Strange isnt it !! finally wearing out the PC

I can't complain about mine, it's over 4 years old and seen use every day and been overclocked, if anything is going it'll probably be either a stick of RAM which won't be expensive or the board, hopefully I can get it to go as far as Kaby or maybe Coffee.

Combat squirrel · 29 Oct 2016 at 23:04

Ok this is weird just changed 1 ram latency settings which its certified for (27 to 28), rest the same, just ran fine for 3 hours, no problems..............confused, but not complaining - yet the ram runs fine at 27 on a different pc.

Doogles · 30 Oct 2016 at 11:25

Combat squirrel said:
Ok this is weird just changed 1 ram latency settings which its certified for (27 to 28), rest the same, just ran fine for 3 hours, no problems..............confused, but not complaining - yet the ram runs fine at 27 on a different pc.

Probably IMC or motherboard then.