Question New PC build frequently freezes, black screen and orange motherboard light ?

[lotlop1]

I recently built my dream PC, this is my second ever build, the first being 10 years back. In everyday use it runs beautifully. But more than once per day the following will happen randomly:

1. my screen will freeze for a few minutes, unable to control keyboard and mouse (sometimes my screen will go straight to black)
2. then my screen will turn black whilst the PC is still running and I will see an orange light on my motherboard (not flashing).

This so far has only happened when I am using the machine casually like browsing or looking through files (then again I have only used this PC for 2 days now). I can only break out of it but long pressing the power button to turn it off. Could this be a RAM issue, PSU, or something entirely different?

CPU temps idle at around 46C and GPU around 45C (using HWiNFO64 for measurement)

Additional info:
1. Not sure if important, but when turning on the PC my motherboard first briefly shows a red light, followed by a white light then no light
2. I have already made sure to download all the most recent CPU, motherboard and GPU drivers
3. I am fairly new to PC building so might need a bit more explaining )

Here are my specs:

Case	FormD T1 2.5
PSU	Corsair SF750 Platinum
GPU	Zotac NVIDEA RTX 4070ti
CPU Cooler	Coolermaster Atmos 240 AIO
CPU	AMD Ryzen 7 78003D
MOBO	ASUS ROG STRIX B650E-I GAMING WIFI
RAM	DDR5 2x32GB 5600MT/s CL40 DIMM
SSD	Samsung 980 PRO 2TB SSD
OS	Windows 11

TLDR: New PC randomly crashes when either idle or doing even the most minimal tasks

 

[faalin]

You need to take a close look at the 4 leds right next to the 24 pin power connector, they will tell you what is wrong with the computer. Reference page 1-21 in you manual to see what each led is and the color.


From the sounds of it, it should be showing a ram issue. Do you know if the ram you bought was on the QVL for the board? You can run Memtest on the ram to see if its ok, there should be no errors, if you get one take the ram back to were you bought it or send it in for RMA to the manufacture.

 

[lotlop1]

Thanks for reaching out! Had a look at the manual and seems the 'Yellow' light is related to the DRAM. I also look on the QVL for my MOBO model and found it Listed here, so it shouldn't be a compatibility.

I'm gonna run a Memtest now and see if I can get any conclusive answers

 

[JasonovichClone]

I recently built my dream PC, this is my second ever build, the first being 10 years back. In everyday use it runs beautifully. But more than once per day the following will happen randomly:

1. my screen will freeze for a few minutes, unable to control keyboard and mouse (sometimes my screen will go straight to black)
2. then my screen will turn black whilst the PC is still running and I will see an orange light on my motherboard (not flashing).

This so far has only happened when I am using the machine casually like browsing or looking through files (then again I have only used this PC for 2 days now). I can only break out of it but long pressing the power button to turn it off. Could this be a RAM issue, PSU, or something entirely different?

CPU temps idle at around 46C and GPU around 45C (using HWiNFO64 for measurement)

Additional info:
1. Not sure if important, but when turning on the PC my motherboard first briefly shows a red light, followed by a white light then no light
2. I have already made sure to download all the most recent CPU, motherboard and GPU drivers
3. I am fairly new to PC building so might need a bit more explaining )

Here are my specs:

Case	FormD T1 2.5
PSU	Corsair SF750 Platinum
GPU	Zotac NVIDEA RTX 4070ti
CPU Cooler	Coolermaster Atmos 240 AIO
CPU	AMD Ryzen 7 78003D
MOBO	ASUS ROG STRIX B650E-I GAMING WIFI
RAM	DDR5 2x32GB 5600MT/s CL40 DIMM
SSD	Samsung 980 PRO 2TB SSD
OS	Windows 11

TLDR: New PC randomly crashes when either idle or doing even the most minimal tasks

I know it sounds obvious but have you check that the GPU is fully rested in the socket.
I suspect it isn't the above, so moving on.

The motherboard traffic lights has something to do with the posting stage, it goes through the sequence of lights and after green it boots up, so it seems like it's related to that. If the GPU or the DDR5 isn't seated properly that could stop it from booting or if you've messed with the XMP setting in the BIOs it could be overclocking beyond it's limits, so I would suggest; go into the BIOS and F5 the settings to default, see if this works.

Failing that, flash your BIOs to the latest update. I know, my new PC built was 5 years since my last one and I had to remind myself all the basics.

The best way to flash is to extract the file to a flash drive make sure it is 16/32 Fat format. When you enter the BIOs go to the EZ-Bios option and select the drive.

Good Luck

Update: Ah sorry I've just seen your last comment, yeah looks like the ram.

 

[lotlop1]

I know it sounds obvious but have you check that the GPU is fully rested in the socket.
I suspect it isn't the above, so moving on.

The motherboard traffic lights has something to do with the posting stage, it goes through the sequence of lights and after green it boots up, so it seems like it's related to that. If the GPU or the DDR5 isn't seated properly that could stop it from booting or if you've messed with the XMP setting in the BIOs it could be overclocking beyond it's limits, so I would suggest; go into the BIOS and F5 the settings to default, see if this works.

Failing that, flash your BIOs to the latest update. I know, my new PC built was 5 years since my last one and I had to remind myself all the basics.

The best way to flash is to extract the file to a flash drive make sure it is 16/32 Fat format. When you enter the BIOs go to the EZ-Bios option and select the drive.

Good Luck

Update: Ah sorry I've just seen your last comment, yeah looks like the ram.

Thank you for the quick response! I haven't considered the GPU but I might go have a look just to confirm it's not the culprit

I've since done and noticed a few things:
- I noticed my BIOS was in fact out of date and updated it with a USB stick
- I ran a memtest and found 0 errors
- I dont currently have XMP enabled
- I did look at my RAM details however, and noticed a pretty big gotcha which is probably the main reason why this is happening. If I look at my Kingston Fury box, although there is a sticker that says "intel" and "AMD", there is a sticker saying "Intel XMP Certified". So I've clearly made a big oops by buying an intel stick for an AMD board...

I guess my follow up question would be, is this a must-replace situation on the RAM or is there a way I can make do with this situation and find ways to minimize the amounts of crashes I can expect.

Hoping that this BIOS update hopefully helps and I might play with enabling XMP (or whatever the AMD equivalent is called)

Thank you for the help!

 

[JasonovichClone]

Thank you for the quick response! I haven't considered the GPU but I might go have a look just to confirm it's not the culprit

I've since done and noticed a few things:
- I noticed my BIOS was in fact out of date and updated it with a USB stick
- I ran a memtest and found 0 errors
- I dont currently have XMP enabled
- I did look at my RAM details however, and noticed a pretty big gotcha which is probably the main reason why this is happening. If I look at my Kingston Fury box, although there is a sticker that says "intel" and "AMD", there is a sticker saying "Intel XMP Certified". So I've clearly made a big oops by buying an intel stick for an AMD board...

I guess my follow up question would be, is this a must-replace situation on the RAM or is there a way I can make do with this situation and find ways to minimize the amounts of crashes I can expect.

Hoping that this BIOS update hopefully helps and I might play with enabling XMP (or whatever the AMD equivalent is called)

Thank you for the help!

No worries bud, happy to assist Yes very true, there's no uniformity in DDR5. The timings for Intel designated ram, don't suit AMD.

Happened to me once, that is why I always stick to a brand that I know works for AMD like the Corsair Vengeance and incidentally this is OK for Intel as well. So easy to go wrong, I made the mistake not checking the label on the Corsair Vengeance, which was listed as "Intel XMP Certified". I thought I could make it work!

Got this (DDR4 for AM4), and I had all sort of problems, you know, PC freezing up or not booting properly. I even tried changing the banks and removing one or two sticks of memory. In the end I returned these back to Amazon and purchased the AMD version of the Corsair Vengeance. Everything was hunky dory after that.

Ps. Change the ram, you shouldn't have to mess with it, it should work out of the box.

 

[faalin]

As long as the ram is on the QVL it doesn't matter if its only intel XMP certified.

The ram in my computer, Assu Prime X670-P WiFi - Ryzen 7 7700x - G.Skill Trident Z5 6000MT/s CL30 - Asus Tuff 4070 TI Super OC, is only certified with intel XMP and is just fine. In fact its the only sticks that could do 6000mt/s at CL30 at the time when i bought them.


You need to look at the lights, they are not only boot up sequence lights but also troubleshooting lights.

 

[geofelt]

You were smart to buy a ram kit that was documented in the motherboard ram qvl list.

XMP is an Intel facility that stores the specs needed to run the ram at full performance in the ram kit itself.
EXPO might be the ryzen equivalent.
Many ryzen motherboards can detect and extract the settings from XMP.
Since you are booting at default specs, xmp/expo should have no impact.

In the bios, do not use load optimized defaults. That may try to load xmp or higher boost clocks.
Let everything default.

Since you ran memtest and got zero errors, the ram itself is likely to be ok.


Look in your installed program list and uninstall anything suspicious.
At setup, you may have been invited to install one or more asus apps.
I would uninstall them.
You also might uninstall the nvidia gpu experience.
At least for now.
If you need a gpu update later, then update the drivers yourself.

They are intrusive, monitoring things and who knows what.
Because memtest runs without windows, your hardware is ok.
Something is impacting you after the initial boot.

 

[lotlop1]

No worries bud, happy to assist Yes very true, there's no uniformity in DDR5. The timings for Intel designated ram, don't suit AMD.

Happened to me once, that is why I always stick to a brand that I know works for AMD like the Corsair Vengeance and incidentally this is OK for Intel as well. So easy to go wrong, I made the mistake not checking the label on the Corsair Vengeance, which was listed as "Intel XMP Certified". I thought I could make it work!

Got this (DDR4 for AM4), and I had all sort of problems, you know, PC freezing up or not booting properly. I even tried changing the banks and removing one or two sticks of memory. In the end I returned these back to Amazon and purchased the AMD version of the Corsair Vengeance. Everything was hunky dory after that.

Ps. Change the ram, you shouldn't have to mess with it, it should work out of the box.

As long as the ram is on the QVL it doesn't matter if its only intel XMP certified.

The ram in my computer, Assu Prime X670-P WiFi - Ryzen 7 7700x - G.Skill Trident Z5 6000MT/s CL30 - Asus Tuff 4070 TI Super OC, is only certified with intel XMP and is just fine. In fact its the only sticks that could do 6000mt/s at CL30 at the time when i bought them.


You need to look at the lights, they are not only boot up sequence lights but also troubleshooting lights.

Ok good to know, spoke to Kingston technical support and got the same answer so shouldn't be too much cause for concern. I'm gonna open my case so that I can see the exact Motherboard light that lights up yellow

 

[lotlop1]

You were smart to buy a ram kit that was documented in the motherboard ram qvl list.

XMP is an Intel facility that stores the specs needed to run the ram at full performance in the ram kit itself.
EXPO might be the ryzen equivalent.
Many ryzen motherboards can detect and extract the settings from XMP.
Since you are booting at default specs, xmp/expo should have no impact.

In the bios, do not use load optimized defaults. That may try to load xmp or higher boost clocks.
Let everything default.

Since you ran memtest and got zero errors, the ram itself is likely to be ok.


Look in your installed program list and uninstall anything suspicious.
At setup, you may have been invited to install one or more asus apps.
I would uninstall them.
You also might uninstall the nvidia gpu experience.
At least for now.
If you need a gpu update later, then update the drivers yourself.

They are intrusive, monitoring things and who knows what.
Because memtest runs without windows, your hardware is ok.
Something is impacting you after the initial boot.

Good idea thank you! I think I might be closer to an answer now. I tried re-seating the RAM earlier today but no change. But I've been looking through the Event viewer for the system logs timestamped around the time of the crash. I can see there are 'critical' errors that give me this 278 BugcheckCode each time. If im not wrong, these relate to a system service or driver correct? so I think you might be right about this being a software issue and not hardware

Here is one of the event logs

Im gonna uninstall the nvidia GPU experience and all the ASUS dross that was installed on setup and see if anything changes and look a bit more into my drivers again and see if there is anything that looks outdated...

Wish me luck!

 

[faalin]

Event 41 bugcheckcode 278 is just a hard hang. Im more interested in the 2 errors just before the critical error.


Anything event 41 is pretty much an effect of the issue, ether a hard hang or system shut down unexpectedly. you need to find the problem that caused the the critical event 41 error

 

[lotlop1]

Good idea thank you! I think I might be closer to an answer now. I tried re-seating the RAM earlier today but no change. But I've been looking through the Event viewer for the system logs timestamped around the time of the crash. I can see there are 'critical' errors that give me this 278 BugcheckCode each time. If im not wrong, these relate to a system service or driver correct? so I think you might be right about this being a software issue and not hardware

Here is one of the event logs

Im gonna uninstall the nvidia GPU experience and all the ASUS dross that was installed on setup and see if anything changes and look a bit more into my drivers again and see if there is anything that looks outdated...

Wish me luck!

Ok it seems like after uninstall the nvidea GPU experience, reinstalling GPU drivers and unsintalling the ASUS apps, my system is running much more stable!

Looking at the event viewer, I had 24 critcal/error events yesterday alone vs only 1 error today which upon inspection was not a dramatic one

TLDR: Seems like uninstalling all this manufacturer software did the trick

will monitor the next few days and see what happens/ and possibly escalate to ASUS if I get more issues

Thank you all