• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

2700x Temperature Reading Anomaly (-10, -57, +107 & +196)

So the rig has been up and running all evening on the 670 now with no hard crash so the aorus 1080ti is not out of the woods yet. Going to leave it on just on the desktop over night as at the moment it seems the hard crash could well be down to the gpu as the temp anomalies never seemed to directly cause the hard crash (during this evening I have seen 5 temp anomalies).

If it's still running by morning that will be the longest I've ever been able to have it running without hard crash which could indicate the two issues were never actually connected. If that's the case I may need to get OCUK to RMA the GPU and whichever component I can narrow the temp anomalies down to.

So for now the RAM / PSU / Case testing will have to wait till morning.
 
So it's morning and the rig has been running all night without the hard crash, just running Kerbal for another hour to see if that forces it but it's looking like the 1080ti is at fault for the worst of the two issues... gutted.

Will carry on testing today with RAM and software, then if I feel up to it I will begin to remove everything from the case to try and pin down the temp anomalies albeit im not as concerned about these as the hard crash!
 
JAW_Jontain said:
So it's morning and the rig has been running all night without the hard crash, just running Kerbal for another hour to see if that forces it but it's looking like the 1080ti is at fault for the worst of the two issues... gutted.

Will carry on testing today with RAM and software, then if I feel up to it I will begin to remove everything from the case to try and pin down the temp anomalies albeit im not as concerned about these as the hard crash!
Click to expand...

If the temp anomalies and the crashing aren't related, you'll probably find that using alternative monitoring software solves the temperature issue.

Keep in mind that if the hard crashing is a result of your PSU then the 670 may not be enough to stress it vs the 1080ti. I know that doesn't help since it adds an extra point of uncertainty, but it is worth keeping in mind. Do you have another psu you can use to test with the 1080ti before sending it back?
 
Had a read through the thread and I couldn’t see what PSU you are running.

You’ve had an amazing amount of bad luck with your build. :(
 
HRL said:
Had a read through the thread and I couldn’t see what PSU you are running.

You’ve had an amazing amount of bad luck with your build. :(
Click to expand...

The PSU is a Corsair RM750x iirc from the other thread. That would make it a single rail unit so it shouldn't be tripping any ocp.
 
Thanks for the replies guys, the PSU is a Corsair RM750x and unfortunately the only other PSU I can get access to is only old cheap 650w which is borderline being too small and don't want to risk it's quality.

I am really hoping that OCUK can replicate the issue when they get the GPU back as this will hopefully help me rule out the PSU and vice versa. The sad yet funny thing is this was the first item back only a month after purchase that they sent me an RMA for but was reluctant to send it back only for it to turn out to be something else! I think i have mentioned before I still have pics and video from the 2nd time the hard crash happened back in Sept 2018 on my phone along with a picture of the rigs first boot where you can see the not so great quality of the cooler fins from the factory.

I hope this is nearing it's end, I've wanted a decent gaming rig since I first ever built a PC and always aspired to have it in a nice Corsair case and for once be able to play games in max settings! 4-5 months in now and I'm desperate to just have a stable rig!
 
welshrat said:
My suspicion would be the power supply.
Click to expand...
Many thanks for the input, what would make you think PSU over the GPU? I called OCUK technical to ask for a way to test the PSU without having a spare but they could only suggest a test which confirms it turns on.

Not going to rule it out but just wondering if there is something you have read in my findings that makes you think that or is it just a hunch?
 
JAW_Jontain said:
Many thanks for the input, what would make you think PSU over the GPU? I called OCUK technical to ask for a way to test the PSU without having a spare but they could only suggest a test which confirms it turns on.

Not going to rule it out but just wondering if there is something you have read in my findings that makes you think that or is it just a hunch?
Click to expand...

Its only a hunch to be honest, but it would always be the first thing I would want to rule out with regard to the hard crashes as it seems exactly the type of scenario a duff power supply would cause issues (i.e when the 1080 needs power), I cant answer with regard the negative temps other than to say my Vega 56 records negative temps in afterburner quite regularly, So I have stopped using anything other than hwinfo
 
Interesting to read about another AMD product recording odd temps, I honestly thought I was the only one! It is rare to see them in Gpuz but have spotted them once or twice. Don't believe I have ever seen them in HWmonitor or HWinfo64 mind.

Have just finished testing the RAM, got my hopes up quite q bit as whilst testing the second stick it took well over an hour before I saw any odd readings but then i registered 2 negative readings and 1 +198. As always the system didn't seem to react in any way to these anomalies and has ran all morning without the crash.

I'm done for today so I will see about getting the GPU sent back and if OCUK cannot replicate the hard crash issue in any way (Going to ask them to thoroughly test as don't mind waiting at this stage) then it must be the PSU/PSU cables or case. There's nothing left to test!

Depending on their response I can move onto building it up outside the case on my next day off on Tuesday.
 
JAW_Jontain said:
Interesting to read about another AMD product recording odd temps, I honestly thought I was the only one! It is rare to see them in Gpuz but have spotted them once or twice. Don't believe I have ever seen them in HWmonitor or HWinfo64 mind.

Have just finished testing the RAM, got my hopes up quite q bit as whilst testing the second stick it took well over an hour before I saw any odd readings but then i registered 2 negative readings and 1 +198. As always the system didn't seem to react in any way to these anomalies and has ran all morning without the crash.

I'm done for today so I will see about getting the GPU sent back and if OCUK cannot replicate the hard crash issue in any way (Going to ask them to thoroughly test as don't mind waiting at this stage) then it must be the PSU/PSU cables or case. There's nothing left to test!

Depending on their response I can move onto building it up outside the case on my next day off on Tuesday.
Click to expand...

Pretty sure only hwinfo can be trusted with Ryzen, everything else can show anomalies, I don't think these temps are of any relevance to be honest (have you seen anything in hwinfo that is negative ?). Could well be the GPU but my money is still on the PSU.

Good luck regardless
 
Last edited:
welshrat said:
Pretty sure only hwinfo can be trusted with Ryzen, everything else can show anomalies, I don't think these temps are of any relevance to be honest (have you seen anything in hwinfo that is negative ?). Could well be the GPU but my money is still on the PSU.

Good luck regardless
Click to expand...
Thanks for the support!

Have been running Kerbal for an hour and a half now with only hwinfo running and not a single temp anomaly showing yet. Feeling more confident about the issues not being linked but will be very interested to see how OCUK get on with the GPU. Half tempted to just order another PSU to test with but will be a big waste of money if that's not it!
 
So at 4 and a half hours running with HWinfo open and No odd readings I am certain the temp anomalies are just an isshe with certain sensor software and hopefully benign.

Have just emailed the good folks at OCUK about getting the GPU RMA'd and tested. I've said it before but I really hope that's light at the end of this long dark tunnel!

Thanks for your advice and suggestions all, hopefully some good news soon!
 
The GPU is now boxed up and on its way back to OCUK so should hopefully hear in about a week how they have got on testing it.

Whilst using HWinfo64 i have noticed one reading that stands out and that is that Chassis2 fan will randomly drop to 0 rpm for a split second before returning to normal? Have just run a quick test to see if this happens at the same time as the temp anomalies in Ryzen Master but it doesn't seem to, none of the other fans are reporting any similar sudden drops in rpm as per this particular fan. When I finally can confirm if its the GPU or PSU causing the hard crash I will be swapping out some of the fans for an AIO cooler setup and needles to say the ML 120 fan that is reporting this issue will be removed from the system.

So a bit more oddness out of this rig, however still yet to have the hard crash since swapping the gpu... (fingers crossed!)
 
Just had a look at the screenshot I got from Sunday after a 4 1/2 hour test with no temp anomalies showing in HWinfo64 and it looks like a few of the fans had reported 0 RPM readings for a second, reading online it looks like a few others have had this issue and its something to do with the sensor not receiving a reading from the fans themselves and doesn't look like its much to worry about...

Think I'm seeing / looking for issues where there are none atm!

However for the more advanced users amongst you is there anything in the power readings that would cause concern? The only one that seems furthest of is the +12v reading of 11.772 but I don't know if that would be considered within tolerance.

hwinfo_pass_by_jontain_dczjxam-fullview.jpg
 
Just noticed this evening that HWinfo64 has a setting that allows displaying for negative temps (for extreme overclocking) that is by default switched off, have enabled this now and going to see if it does actually register any of the temp anomalies as the negative readings are much more frequent than the higher positive readings, hopefully it wont but am going to do some further testing tonight just to be sure.
 
No negative readings so far thankfully.

However this evening whilst considering swapping out the standard Wraith Prism cooler for a h150i pro I have been playing with some online PSU calculators and am now doubting if the Rm750x is actually enough for my rig...

If any of you can weigh in on the post I have put up in the power supply section I would greatly appreciate it:
https://forums.overclockers.co.uk/threads/rm750x-enough.18845995/
 
Update from OCUK after testing the GPU and has apparently passed without issue. Going to order a new PSU tonight from overclockers and see if my brother can get an RMA. Will change all psu cabling once it turns up.

Going to also go from RM750x to RM850x to bed 100% it's got enough power.
 
JAW_Jontain said:
Update from OCUK after testing the GPU and has apparently passed without issue. Going to order a new PSU tonight from overclockers and see if my brother can get an RMA. Will change all psu cabling once it turns up.

Going to also go from RM750x to RM850x to bed 100% it's got enough power.
Click to expand...

Good luck, I can highly recommend BeQuiet (My 630 watt version happily powers my 2600 an vega 56) as an alternative if that takes your fancy, real shame you cant borrow one from somewhere just to be sure. To be honest though PSU would have been my first port of call.

Anyhow you must be close now to a resolution.

Cheers Sam
 
Thanks for the reply Sam! Am really hoping so as the only things I that havnt been changed are the case, the m2 drive and the RAM (ram passed all suggested tests).

Must have just been unlucky with the PSU and it only causes the crash issue with the power requirements of the 1080ti as you guys suggested, it has been stable with the 670 for the last week so will update as soon as I get the PSU and associated cables replaced and hopefully will have finally nailed it.

Just want to say I am really thankful for all OCUK's help with this along with the great support and suggestions from this community!
 
You must log in or register to reply here.
Back
Top Bottom