I9 9900k

moorhen2 · 22 Nov 2018 at 18:16

This is a 5.0ghz 15 minute RealBench stress test run, under load vcore is 1.154v, passed 15 minutes, but it could well have crashed in the next minute, I could run it for 8 hours like we used to, and say it is stable, but it might have crashed after 8 hours, and so it could go on, where do you draw the line. ??

Sc00p007 · 22 Nov 2018 at 18:43

moorhen2 said:
The thing is can we ever say our OC is fully stable, there are many things that can come into force that will cause your system to GSOD, its not just an unstable OC, drivers can be a major cause of crashes, W10 is quite a stable OS, but it only takes something to happen it doesn't like and it will crash, and its not always the hardware that causes it.

Yes, I agree and that's why I said head scratcher. The fact that bsod or system shutdown can be caused by the other factors you list, is why it has frustrated me in the past because my oc was quite low down on the list. Also with it being maybe once a month, it was a long drawn out itch. I now tend to go with a few stress tests of varying intensity and type and run them for 4-5hrs each. My gaming session time being why I chose 4-5hrs. Since then any rare bsod has always been driver, other hardware etc.
However, the main reason why I have been vocal in this thread is now i'm unsure of prime95avx. This has always been in my suite and proved to be very good for me over the years. I've also been able to oc without offset with my chips. Now I can't

So my chaos and order are having a chat. I want to drop prime now, but then I will have an itch. Need someone to convince me that it is indeed a "virus"

.

moorhen2 · 22 Nov 2018 at 18:57

I think you already know my views on P95, why on earth people inject so much heat into there Silicon is beyond me, but each to there own I suppose. I know there is a non AVX version available, but it still heats up the chip to degrees that the everyday user would never subject there cpu to.

Distracted · 22 Nov 2018 at 22:30

On the subject of WHEA, do these occur with stock CPU settings?

What i mean is are there a number of recoverable or correctable errors in everyday use anyway?

Distracted · 23 Nov 2018 at 16:34

Distracted said:
On the subject of WHEA, do these occur with stock CPU settings?

What i mean is are there a number of recoverable or correctable errors in everyday use anyway?

Seems like people only notice these in everyday use when they cause unrecoverable errors.

On another note I do not have everything i need for my loop so i haven't got back to the 9900k testing. Ocuk saturday delivery to the rescue!

moorhen2 · 23 Nov 2018 at 17:12

Distracted said:
Seems like people only notice these in everyday use when they cause unrecoverable errors.

On another note I do not have everything i need for my loop so i haven't got back to the 9900k testing. Ocuk saturday delivery to the rescue!

What loop have you got my friend, if I may ask. ?

Distracted · 23 Nov 2018 at 17:19

moorhen2 said:
What loop have you got my friend, if I may ask. ?

Right now it is ek soft tubing with phanteks block on the 1080ti, ek coolchill(?) built in block on the maximus formula xi vrms, ek velocity on the cpu and ek xres 140 and pe360 rad.

I'm going to be switching to hard tubing and possibly swapping the pe360 for a tx360 to give me more space in the case, as well as adding a mo-ra3 360 pro.

I'm doing it primarily for noise reduction and looks since the pc is being wall mounted. I've a lot going on at the moment so building the pc has been a long drawn out process. Everytime i think i can stop and get it done something else comes up.

Hopefully this weekend will be the one.

Distracted · 23 Nov 2018 at 18:45

I forgot to mention that I'm adding the mo-ra via soft tubing and qdcs as well as adding temp sensors and a flow meter to the main body once i get it all apart.
I need more loop info and i have to set up better control of the fans in response to temp changes as well.

moorhen2 · 23 Nov 2018 at 18:55

Distracted said:
I forgot to mention that I'm adding the mo-ra via soft tubing and qdcs as well as adding temp sensors and a flow meter to the main body once i get it all apart.
I need more loop info and i have to set up better control of the fans in response to temp changes as well.

Sounds good, I have an xspc Raystorm Pro block, xspc dual pump/res, 2 xspc 360's, an xspc 240 and xspc 120 inside my case. Cools really well.

Distracted · 23 Nov 2018 at 18:58

moorhen2 said:
Sounds good, I have an xspc Raystorm Pro block, xspc dual pump/res, 2 xspc 360's, an xspc 240 and xspc 120 inside my case. Cools really well.

I can't wait to get mine sorted. It is reading about and seeing other people's setups that has left me poor and surrounded by waiting parts

I've got a ddc pump as well that i might add on the mo-ra if i see the flow rate dropping, but that will mean another delay. Right now i'm hoping the 1 d5 will be enough but with 4 qdcs and a massive rad I wont know until i try it.

0121danwilliams84 · 23 Nov 2018 at 21:18

Distracted said:
On the subject of WHEA, do these occur with stock CPU settings?

What i mean is are there a number of recoverable or correctable errors in everyday use anyway?

Hey dude. So, I got a chance to tinker again. 1.340v llc changed from 5 to 6 maximum voltage has now gone up to 1.376v. Seems level 6 is a lot more deadly! Still got a WHEA ffs. I wonder if I bother 1 notch higher vcore. Probs go to 1.4v max then give up. It's not crashing it just giving the errors in HW Info.

Distracted · 23 Nov 2018 at 21:24

I don't know how critical the errors are. Do they build up, getting worse over time, or are they isolated incidents that are corrected and logged but otherwise have little impact on performance and stability.

I was reading reddit and a guy was saying he sees whea errors but his system is stable for his games and whatnot so he just left the voltage as is. I imagine a lot of people do without realising they are getting errors. After all if you are going for a quick and easy oc then you set, test if whatever you want to run works, then forget.

I wonder if the auto oc motherboards apply produce errors. I doubt anybody is checking for them. Then again they tend to be heavy on the voltage so perhaps not.

Distracted · 23 Nov 2018 at 21:27

0121danwilliams84 said:
Hey dude. So, I got a chance to tinker again. 1.340v llc changed from 5 to 6 maximum voltage has now gone up to 1.376v. Seems level 6 is a lot more deadly! Still got a WHEA ffs. I wonder if I bother 1 notch higher vcore. Probs go to 1.4v max then give up. It's not crashing it just giving the errors in HW Info.

Have you set ram to stock to make sure the errors are related to the cpu oc?

0121danwilliams84 · 23 Nov 2018 at 21:28

Distracted said:
I don't know how critical the errors are. Do they build up, getting worse over time, or are they isolated incidents that are corrected and logged but otherwise have little impact on performance and stability.

I was reading reddit and a guy was saying he sees whea errors but his system is stable for his games and whatnot so he just left the voltage as is. I imagine a lot of people do without realising they are getting errors. After all if you are going for a quick and easy oc then you set, test if whatever you want to run works, then forget.

I wonder if the auto oc motherboards apply produce errors. I doubt anybody is checking for them. Then again they tend to be heavy on the voltage so perhaps not.

I usually get about 1 or 2 'WHEA Error 19' in most games but the game just continues as normal.

Yeah, I agree. I am sure it means unstable but I bet loads of people set and forget and don't even look for errors only blue screens.

0121danwilliams84 · 23 Nov 2018 at 21:29

Distracted said:
Have you set ram to stock to make sure the errors are related to the cpu oc?

The Rams been on XMP the whole time. I messed with the SA and IO settings but I put them back to auto to eliminate that.

Distracted · 23 Nov 2018 at 21:38

Have you googled whea error 19?

Seems to be an issue either related to ram or spectre and meltdown patches or both.

It might have nothing to do with your oc.

I could be wrong but those are the common results i saw when searching.

Distracted · 23 Nov 2018 at 21:43

In other news I've got confirmation that my order from OCuk is arriving tomo.

:cool:

pretty excited if I'm honest. Please don't mess this up for me DPD.

All being well I'll have my PC running again tomo evening with an extra rad and some hardline.

0121danwilliams84 · 23 Nov 2018 at 21:43

Distracted said:
Have you googled whea error 19?

Seems to be an issue either related to ram or spectre and meltdown patches or both.

It might have nothing to do with your oc.

I could be wrong but those are the common results i saw when searching.

I saw those results. I also saw people saying cpu overclock and vcore sorted it. I can't keep increasing forever tho.

Distracted · 23 Nov 2018 at 21:44

0121danwilliams84 said:
I saw those results. I also saw people saying cpu overclock and vcore sorted it. I can't keep increasing forever tho.

Turn off xmp and see if you are still getting them with the oc you have now.

Distracted · 24 Nov 2018 at 08:32

0121danwilliams84 said:
Hey dude. So, I got a chance to tinker again. 1.340v llc changed from 5 to 6 maximum voltage has now gone up to 1.376v. Seems level 6 is a lot more deadly! Still got a WHEA ffs. I wonder if I bother 1 notch higher vcore. Probs go to 1.4v max then give up. It's not crashing it just giving the errors in HW Info.

I thought this might interest you. If i am reading it right it is a user's results for oc a 9900k on a z370 hero and air cooler.

https://www.overclock.net/forum/6-i...0-vrm-discussion-thread-335.html#post27729352