• 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)

Thanks for the suggestion but I don't have a test bench or anyway of building out of the case?

I have read online about a lot of people having freezes/crashing that relate to Ryzen voltage/power plans which could make sense for the hard crashing issue. Something to do with it going from low clock to higher clocks but getting insufficient voltage?

I think my plan is to let it finish memtest86 with all 4 passes to ensure it's not RAM related. Get ocuk to have a look and hopefully replace the cpu as not reading about anyone having issues with these random temp readings!

Once that is done if the hard crashes persist I will have to see if I can change the power plan or set the cpu cores not to ramp up and down (will prob need some assistance as to what to change in bios?) which will be unfortunate as this is something that they are designed to do. Finally if none of that works I can try building it out of the case but not quite sure how id safely do this.
 
Interesting theory for hard crash:
I had tried a lot of things since the original build to work out the hard crash and thought wrongly that getting rid of ai suite had done the trick... However another thing I had done was change the power plan to performance, albeit windows like to reset this to balanced, but the crashing went away for a good while even though the temp anomalies remained.

The hard crashes only came back when trying to diagnose the temp anomalies and this would've been after installing the latest AM4 chipset drivers as recommended which aparantly auto changes the power plan to a ryzen specific one. Reading online a few people have had issues and have actually been recommended by AMD to set this to the normal windows balanced plan instead.

Could be coincidence that the hard crashes came back with chipset driver update but once memtest86 is done with all 4 tests I am going to boot up and check what power plan is selected. If it is the Ryzen one I am going to try the windows plan to see if I can leave it downloading without the hard crash. I suspect it won't fix the chips temp anomalies and they could still be related but thought it's worth a try!
 
Last edited:
Ok so Memtest86 ran all 4 passes without issue which is great, should mean at least the memory is stable.

There is an update for Ryzen Master so just getting this done as well as finding the chipset driver installer on AMD's site without Ryzen Master included so wondering if I hadn't actually updated to the latest chipset driver when installing Ryzen Master so going to get that done as well.

In the meantime the power plan was set to Windows balanced as no Ryzen power plan was showing so have put this to Performance and have noticed straight away that all the cores seem to be locked between 4025-4050mhz without the fluctuations showing on the balanced plan. Going to run the rig for a while on this setting and see if the temp anomalies / hard crashes are still present.
 
Ok feeling somewhat silly...

It would appear that the drivers suggested for Ryzen 2700x installs Ryzen Master alone and it does not in fact install the latest chipset driver...

Now when using the chipset driver installer it didn't actually detect any drivers installed for the chipset, once installed I could see the Ryzen Balanced power plan so have switched to this, will check windows doesn't default the power setting back to balanced on reboot but I am going to need to do some more testing now I actually have some chipset drivers installed... doh!

EDIT: Just restarted and it retained the Ryzen Master power plan but still all cores sitting at around 4000mhz rather than jumping around from 2200-4050 mhz at idle which I thought was a key feature of the X series chips... will do some testing with this current drivers and will owe the guys at OCUK an apology if that has fixed it.

EDIT2: Changing the power plan back to windows balanced immediately set the cores jumping around as before and intended, going to test on this setting to see if either the temp anomalies or hard crashes persist.
 
Last edited:
Ah never mind, just recorded a -58 reading even with the latest version of Ryzen Master / Chipset drivers.

Just need to see if these updates help with the hard crash issue, going to leave it idling whilst downloaded as that has caused the hard crash pretty consistently.
 
Thought I would test how it performs with the Ryzen power plan this evening and whilst I didn't record any temp anomalies during a short stint in Kerbal I decided I would finally play a bit of RE2 and unfortunately got a hard crash after about 10-15 mins.

Going to email OCUK now with this weekend's findings and get the CPU sent off tomorrow, keeping my fingers crossed that this will solve the issues!
 
So today is the day my replacement CPU has arrived from the good guys at OCUK... was feeling hopeful so installed the new cpu.

Left RAM at stock settings to just be safe although this did pass the tests at 3200mhz with no issues.

Unfortunately after 10 long minutes just on desktop letting programs download updates and Ryzen Master running I have already got a reading of -10... No hard crash as of yet but I am just waiting for it now.

I feel sick.
 
This leaves GPU, PSU, RAM, Case or Hard drives left and I don't know where to even start.

For some reason installing the new video card drives also closed Ryzen Master...
 
JAW_Jontain said:
So today is the day my replacement CPU has arrived from the good guys at OCUK... was feeling hopeful so installed the new cpu.

Left RAM at stock settings to just be safe although this did pass the tests at 3200mhz with no issues.

Unfortunately after 10 long minutes just on desktop letting programs download updates and Ryzen Master running I have already got a reading of -10... No hard crash as of yet but I am just waiting for it now.

I feel sick.
Click to expand...

Ouch. That’s a pain the backside.

Tried a clean install and not bothering with Ryzen Master? Most tweaking can be done in the BIOS anyway.
 
Thanks for the reply, starting to formulate a plan going forwards. This evening I have tested and ruled out (i.e. still getting bad readings):-

- Different wall socket and surge protectors (No change)

- Removed and unplugged both speakers and headphones (No change)

- Tried to use on board graphics until I realised that Ryzen does not support this so cannot rule out gpu just yet

- Removed all pci slot covers in case they were causing an issue by contacting edge of mb (No change)

- Tested with all hd unplugged but the m2 drive with windows on and took a while but then got a anomalie at 45 mins (oddly very shortly after closing case door...)

Out of time for testing tonight but will be trying :-

- Case doors off, either side and both (likely this was coincidence but got three bad readings shortly after closing the hard drive side door)

- One stick of ram at a time

- Removing software one at a time and then starting with fresh install and reverse

- Getting hold of a different but likely older gpu to test.

- Remove case connectors one at a time before attempting to build outside of case

- Finally that will leave PSU/cables however I have absolutely no idea how i would test this.

I know some of these tests sound odd but really at a loss as to what to do so trying to rule out everything... Thankfully no hard crash as of yet which is the worse of the two issues but atm just trying to work through it one by one starting with the easiest things first!

The saga continues...
 
Last edited:
Thanks for the reply, I appreciate the feedback. Once I know the rig is stable I will be using hwinfo64 for monitoring if I do ever decide to do any overclocking however ryzen master is the only software I have found so far that displays these anomalies consistently and rightly or wrongly Ive believed the hard crash would be linked to this.

But I will add this to the list of tests to see if it prevents the hard crash although I have had it without any monitoring software running.
 
Had a response from OCUK now suggesting to send back the RAM however I am not sure it was from one of their guys that is familiar with the issue as suggested trying Memtest86 so have said I need a bit more time to try my best to rule out everything I can before sending any more items back.

Hopefully got hold of an older nvidia card to drop in to try over the weekend which means I can test if the gpu is causing any issues. Will also try one stick of ram at a time to see if this is anything to do with it although with both sticks they passed Memtest86 with no problems.

Then will be onto weirder issues to do with case or possibly the PSU, if anyone can suggest a way of testing the PSU I would greatly appreciate this!
 
Decided after a long day i didnt want to spend my evening testing tonight so thought i would try some gaming. Managed to get to play on the rig for most of the evening without issue but the hard crash just happened and this was without any monitoring software running in the background. Unfortunately this confirms both issues are still present so will be spending Sunday carrying on with the testing GPU and RAM to see if this brings any results.
 
Last edited:
Assuming it is NOT a software issue, you could try the following.

Take everything out of the case and set the motherboard on the motherboard box, or another equally tall box making sure there is room for the PCIE bracket of the GPU to hang over the edge. Then add the CPU, GPU, Main drive, 1 stick of RAM and the PSU. Mount the CPU cooler and whatever fan it requires, ensure everything is connected to the PSU and connect your peripherals, keyboard and mouse only and then a single monitor.

Once this is done, see if the problem persists, if it does you've just ruled out the case shorting. Now is the time to swap individual components such as the RAM, GPU, PSU or HDD but only 1 at a time, hopefully this will identify the problem part. If swapping them all doesn't find it (hell swap everything you can, even the peripherals but only one at a time and test) and you've already had the CPU replaced then it can only be the mobo.

If when you set up your PC outside the case with minimal components and it does not display the problem, then slowly add additional components, such as RAM, HDD, Peripherals. One at a time and test for an extended period doing your normal activities and stressing the machine if possible and you have the time. This should reveal what is causing the problem.

The only problem you may have with this setup is turning the PC on if your Mobo does not have an onboard on switch. If this is the case then plug in just the power switch of the case and keep everything else outside the case. It will be awkward and take up space and be a PITA but it will help to isolate the problem.

HTH
 
Thanks for the reply, once I test the GPU and RAM on Sunday I will move onto removing software one at a time. If this yields no results then I will move onto testing out of the case as suggested.

I can rule out motherboard as this was replaced first before the CPU! Starting to think it may be GPU again but have got an older 7 series Nvidia which was stable for many years so will be testing this first on Sunday, will be a shame as this was my first thought months ago when I first encountered the problem but needless to say once I've tested everything individually I should hopefully be able to narrow it down!

EDIT: Only thing that is slightly worrying is the aorus 1080ti doesn't show on overclockers site anymore so if it does turn out to be the GPU I'm not sure how this will be replaced if needed!
 
Last edited:
JAW_Jontain said:
Thanks for the reply, once I test the GPU and RAM on Sunday I will move onto removing software one at a time. If this yields no results then I will move onto testing out of the case as suggested.

I can rule out motherboard as this was replaced first before the CPU! Starting to think it may be GPU again but have got an older 7 series Nvidia which was stable for many years so will be testing this first on Sunday, will be a shame as this was my first thought months ago when I first encountered the problem but needless to say once I've tested everything individually I should hopefully be able to narrow it down!

EDIT: Only thing that is slightly worrying is the aorus 1080ti doesn't show on overclockers site anymore so if it does turn out to be the GPU I'm not sure how this will be replaced if needed!
Click to expand...
I imagine if it IS the GPU and OCUK don't have stock then they can either speak to the parent company and request stock (if any exists) or they will offer you an RMA which will mean sending it off and a replacement Aorus being sent to you. But I would cross that bridge when you get to it, find the fault first, then worry about how it will be resolved. It may not be the GPU.
 
Distracted said:
I imagine if it IS the GPU and OCUK don't have stock then they can either speak to the parent company and request stock (if any exists) or they will offer you an RMA which will mean sending it off and a replacement Aorus being sent to you. But I would cross that bridge when you get to it, find the fault first, then worry about how it will be resolved. It may not be the GPU.
Click to expand...
Yep your right don't mean to get ahead of myself it's just the hard crash only shows a physical change to GPU (lights out and power leds light up) and last night when booting up I got a second of strange screen tearing artefact as it loaded the windows login screen (hope that the hard crashes aren't having a negative impact on the GPU!)

But swapping the card out this weekend should hopefully confirm if this is the case, many thanks for your input!
 
Just done the gpu swap after testing removing Aorus Engine software (no change) and uninstalling all nvidia drivers.

Its actually a GTX 670 not a 7 series card but hopefully this may help me to see if its gpu related or not! Kinda hoping this doesn't solve it cos i really like that 1080ti

My only worry is if it's PSU related it might also not show as the 670 is going to be a lot less power hungry than the 1080ti but we shall see how it goes.

EDIT: First thing I have noticed before the drivers have even installed is that the temp curve in Ryzen Master still jumps up but has much longer periods of lower temps, and again once the graphics driver got install Ryzen Master closed as with last time.
 
Ok so after a short test in Kerbal I am happy as have had a -58 degree reading, the only difference (other than the FPS!) to the 1080ti is with the 670 the temp anomaly showed for 2 seconds rather than the 1 second it was always showing with the 1080ti. Never been so happy to see this issue as I am right now lol.

Going to leave it running for a while now to hopefully get the hard crash so I can fully rule out the 1080ti being at fault for both crashes! Needless to say if I can replicate the hard crash I will carry on testing with the 670 as happier to put this through the crashes than the 1080ti!

The plan after seeing if I can get the hard crash with 670 is to move onto RAM before software and finally building out of the case.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom