Cold boot reset loop.
Cold boot reset loop.
Hello.
So I have an issue I have searched for high and low.
I believe the issue started in the past 2 weeks.
Basically every time, I start my PC from cold(was off overnight) I get this reboot cycle.
I'll attach a video link that I uploaded to youtube on private.
First off, my hardware.
Here's the thing. I have recently upgraded my PC quite a bit.
Full specs
:
Corsair Vengeance LPX 3000mhz memory
(4 years old)
Samsung 850 Evo Sata SSD 500gb (
4 years old
)
Gigabyte z390 ud mobo
(about 2 months old)
Intel 9700k
(about 2 months old)
Kingston A2000 500gb m.2 SSD
(about 2 months old)
Corsair H100i RGB PRO XT AIO
(about a month old)
HX1200i Platinum PSU
(about 2 weeks old)
Gigabyte Gaming OC 8GB RTX 3070
(about a week old)
1920x1080p 24Inch Asus monitor
, very old(4-6 years) with HDMI
2560x1440p 27inch Gigabyte Aorus cv27q
(about half a year old)
I'm not certain if the issue started after the PSU installation or GPU. I'm leaning towards the GPU installation. Since I'm always troubled with things that happen with my baby.
As soon as I discovered it did this, I started searching for solutions.
So next bit of information are
overclocks
.
The only 2 things that I have overclocked are the RTX 3070(with aorus software, furmark 2 hours stable)
GPU
voltage is locked and Power draw is overall extended with the gigabyte version from Nvidia-s TDP 220 to 270w. So that is also locked to the 270w.
My
PSU
comes with a switch, that allows it to switch from single to multirail. (100A vs 40A) I have it on default, which should be single rail.
Memory
is using an XMP profile to reach that 3000 mhz with 1.35 volts.
I have tried both without overclock.
Cpu is on auto even tho I have overclocked it in the past. Nothing crazy, 4.9ghz and Vcore was about 1.28-1.31.
Things I have tried so far
:
Checking the event viewer for any Critical issues, Errors or warning related to the issue. I have found none. Only the 3 warnings that I had after the boot were User Device Registration, DistributedCOM and CertificateServicesClient-AutoEnrollment. Non important issues with certificates or permissions.
Doing a bios reset and hard reset. Disconnected everything. Let out the residual power from the PC via power button. Disconnecting the Bios battery and doing the same thing. Didn't fix the issue(that was done yesterday)
Trying to see if the case power button was forcing a reboot cycle, so I disconnected it, replaced it with reset button and booted via reset buttong. Everything worked the same way and since it's hard to duplicate the issue without waiting overnight or some longer period of time, i'm forced to longer timelines.
Checking all connections leading to the PSU, motherboard and my SATA device. Everything seems to be connected properly.
Running multiple tests like : OCCT system stability with Large and Small data sets for 3-4 hours both. Running memtest86 for 4+1+1 passes , all tests. Running furmark with desired overclock for 2 hours straight, pluss alot of shorter periods of time. All came back clean. No errors.
Tried to reconnect my GPU PCIE cables to the same modular row on the back of my PSU(didn't know if it makes a difference)
Tried reducing the number of devices using the same extension cord. Used to be that my TV and speakers were also connected to the same extension cord but now I'm only running my 2 monitors and PC off of it.
Tried updating drivers with driverbooster and manually for chipset.
Bios update is done a few months back, latest bios installed f10J.
I probably have tried a few things more, but I'm not remembering them currently.
So my question is. How likely is it, that I may have damaged my motherboard in the installation process of my PSU or GPU.
Or is it possible, that my 1200w PSU draws too much power on cold boot for some reason that it makes some sort of protection trigger and it shuts off?
Or I've seen some possible explanations from older posts that some CPU-s and MOBO-s have auto overclock features that do it purposly to find the most stable clock(even tho I haven't OC-d them)
Or if my memory fails to train is it possible that it doesnt show up with memtest86?
Any help is appreciated.
Many thanks
Maidu
Edit2: Here's the new link, might still be loading:
Cold boot cycle.
Update 4. The faulty unit was PSU, it fell into the date range of the faulty hx1200i-s produced between the 30th and 41st weeks of 2020 (July to early October). Serial numbers range from 2030XXXX to 2041XXXX. Fortunately, the RMA remains valid, and a replacement HX1500i is being sent. Hope this assists someone in the future.
Your video link appears as a private video and cannot be displayed. When you begin your computer, observe the codes on the led window and check if they stop when it shuts down. In the meantime, attempt to disable fast startup in your BIOS and quick startup in Windows settings. See if this resolves the issue.
Hey, I'll update it to unlisted but the main thing is the cpu led at first. Then it restarts and shows CPU-dram before resetting again, until everything cycles properly and works normally. No errors in event viewer. No kernel power, nothing. I'm leaning toward either RAM not training or a motherboard issue.
On the same point, I tried putting it to sleep and it behaved the same way—just one restart but it worked.
Yes, that was one of the first things i troubleshooted because it appears to cause many problems on desktop computers linked to drivers. The problem seems to occur even before initializing bios. There is no display output. Fans are spinning at high RPM and the LED strip is turning on. The integrated LED strip with a delay is also visible.
Double channel setup. 2x8gb storage. I've already placed orders for replacement parts and a new motherboard. It seems, as of now, the issue likely lies with the motherboard itself. Since the chassis power button won't activate even though the system is booting normally, the front header connections appear to be the only direct points related to the motherboard.