My new pc down why it happened and how to fix?

[khjhpopasa]

My new pc down why it happened and how to fix?
I bought a new pc spec
My-Pc: R5 7600X, RX 6650 XT(I return my new gpu 7900xtx), B650 Aorus Elite AX, DDR5 6400MHz 16x2 cl32, SSD , Corsair RM 1000e.
the problem is when i start my pc it work all the time if i watch youtube video or programming it work perfectly
the problem start when I start to play (gamming)

When i play the game start and then i can play 5min-90min something between the pc turn off all display ports fan and anything like he off.
But the pc still on I see motherbaord light and click button light still on, but all another thinks just off without sound and nothing (keyboard & fans & fans RGB iclude case and cpu fans & etc all off).
if I connect usb the light is on only on usb.
If I click off button nothing happened even after long click.
only one way to turn off that pc with psu button to off (0) how I should fix that issue ? why it happened ? and how to find the problem?

The weird is I can play sometimes and then it happened can be 5 min game or even more then hour.
Please help me and thanks!!

 

[natr0n]

who built your pc? what are your temperatures?

 

[khjhpopasa]

who built your pc? what are your temperatures?

I build it and on full load temp is fine.
GPU 70C cpu 90C on full load

 

[natr0n]

Your cpu is the problem. YOu need better cooling. It will thermal throttle at that temp and then shut off system eventually when it spikes too high.

check bios see if you have eco mode that should fix that

 

[Vya Domus]

Your cpu is the problem. YOu need better cooling. It will thermal throttle at that temp and then shut off system eventually when it spikes too high.

Tjmax for these CPUs is 95C, it should not throttle anywhere close to being problematic even at 90C.

 

[Toothless]

Your cpu is the problem. YOu need better cooling. It will thermal throttle at that temp and then shut off system eventually when it spikes too high.

check bios see if you have eco mode that should fix that

Yeah, no. The new AMD stuff will push harder and hotter than before. OP isn't having a heat issue.

 

[natr0n]

Dude is on the edge of failure within a few degrees and thats fine.... ok you "pros" handle it.

 

[Vya Domus]

Dude is on the edge of failure within a few degrees and thats fine.... ok you "pros" handle it.

This is from TPU's own review, 90C is fine.

 

[agent_x007]

Check cables on both ends (plugs/connectros/headers - if they are all are "locked", and secured in place).
Any places where you have tight bends ?
XMP enabled ?

Basicly, either PSUs overcurrent/overvoltage protection get's triggered (which switches off PC under load), OR your MB switches off PC because of some fault (maybe too high temp on VRMs ?).
I should add that motherboards with SPS style MOSFETs can be switch off by PWM controller itself (regardless of what VRM temps you see in diagostic tools, since not all inputs made by PWM Controler/MOSFETs can be checked in such tools).
Checking VRM's thermal pads may be required.

Lastly : RGB... overheating RGB specific controllers or "mixed" RGB/FAN/LED Strips ones, are also a thing to watch out for.

 

[Toothless]

Dude is on the edge of failure within a few degrees and thats fine.... ok you "pros" handle it.

Don't be salty dude. AMD THEMSELVES have stated the max operating temperatures for their chips and it's been all over tech headlines when the 7xxx series launched. 5xxx like your and my chip is nothing like the 7000 stuff. Apples to oranges.

 

[xtreemchaos]

the 7000s do run hot even 90c is within there spec, you could have a ram problem, try running without xmp turned on.

 

[bug]

Check cables on both ends (plugs/connectros/headers - if they are all are "locked", and secured in place).
Any places where you have tight bends ?
XMP enabled ?

Basicly, either PSUs overcurrent/overvoltage protection get's triggered (which switches off PC under load), OR your MB switches off PC because of some fault (maybe too high temp on VRMs ?).
I should add that motherboards with SPS style MOSFETs can be switch off by PWM controller itself (regardless of what VRM temps you see in diagostic tools, since not all inputs made by PWM Controler/MOSFETs can be checked in such tools).
Checking VRM's thermal pads may be required.

Lastly : RGB... overheating RGB specific controllers or "mixed" RGB/FAN/LED Strips ones, are also a thing to watch out for.

This. What PSU are you using? VRMs aren't pushed too hard while gaming, I think it's more likely with the GPU not getting enough juice under load.
Also check Event Viewer, is that something recorded there?

 

[Assimilator]

Please try to do some basic troubleshooting yourself before posting here.

 

[Jacky_BEL]

Don't assume it is a hardware fault.
As it happens, I just had a PC BSOD 'ing on me 15 to 20 minutes after startup,
or when I tried uninstalling some programs from that PC.

After SFC /scannow I found in the logs that a whole lot of files got corrupted (think it was net framework 4 files).
I had to do a (windows10) repair install to fix it , (using the keep settings and data option).

 

[Eskimonster]

Don't assume it is a hardware fault.
As it happens, I just had a PC BSOD 'ing on me 15 to 20 minutes after startup,
or when I tried uninstalling some programs from that PC.

After SFC /scannow I found in the logs that a whole lot of files got corrupted (think it was net framework 4 files).
I had to do a (windows10) repair install to fix it , (using the keep settings and data option).

BSOD´s can corrupt files

 

[xtreemchaos]

Don't assume it is a hardware fault.

funny thing is yesterday i had a bsod after installing nvidias latest drivers, couldnt do anything the rigg would start get into windows and restart i had to go int safe mode option 5 and remove the drivers the install the older driver and alls good. nothing to relate to the OPs post but id thought id mention it just in case.

 

[khjhpopasa]

Tjmax for these CPUs is 95C, it should not throttle anywhere close to being problematic even at 90C.

But on game my cpu not full load he temp only 58C and pc still get bug

Your cpu is the problem. YOu need better cooling. It will thermal throttle at that temp and then shut off system eventually when it spikes too high.

check bios see if you have eco mode that should fix that

But on game my cpu not full load he temp only 58C and pc still get bug

Tjmax for these CPUs is 95C, it should not throttle anywhere close to being problematic even at 90C.

But on game my cpu not full load he temp only 58C and pc still get bug

the 7000s do run hot even 90c is within there spec, you could have a ram problem, try running without xmp turned on.

But on game my cpu not full load he temp only 58C and pc still get bug.

XMP can do bug?

 

[pavle]

Try updating the system/mainboard BIOS and then running it at setup defaults.

 

[xtreemchaos]

XMP can do bug?

not a bug as sutch but a bad ram stick when being overclocked by XMP can cause BSODs and other sudden shut downs.

 

[c2DDragon]

XMP can do bug?

Yes, especially if your motherboard doesn't support your RAM frequency.

I see there : https://www.gigabyte.com/Motherboard/B650-AORUS-ELITE-AX-rev-1x/sp#sp that it supports 5200Mhz maximum and THEN it is "(OC)", up to 6666Mhz, it means that you need the last BIOS to be able to use 6666Mhz and some older ones for less "(OC)" frequencies but newer than what you get when you just unboxed your motherboard.

Did you update your BIOS before enabling XMP ? 6400Mhz isn't supposed to be supported by the very first BIOS.

Check here : https://www.gigabyte.com/Motherboard/B650-AORUS-ELITE-AX-rev-1x/support#support-dl-bios and notice the "Improve memory compatibility" line, it means you can use higher RAM frequencies each time, now it doesn't tell much but I guess you would need the F3b or F3h BIOS for 6400Mhz.

 

[khjhpopasa]

Try updating the system/mainboard BIOS and then running it at setup defaults.

I have f1 version bios but why do you think to cause a problems because I see on list nothng mention about bug ram in version bios.
Any way I cahnge to 4800MHZ as base and will check for crushes on a game again and if it happned ill let you know.

Yes, especially if your motherboard doesn't support your RAM frequency.

I see there : https://www.gigabyte.com/Motherboard/B650-AORUS-ELITE-AX-rev-1x/sp#sp that it supports 5200Mhz maximum and THEN it is "(OC)", up to 6666Mhz, it means that you need the last BIOS to be able to use 6666Mhz and some older ones for less "(OC)" frequencies but newer than what you get when you just unboxed your motherboard.

Did you update your BIOS before enabling XMP ? 6400Mhz isn't supposed to be supported by the very first BIOS.

Check here : https://www.gigabyte.com/Motherboard/B650-AORUS-ELITE-AX-rev-1x/support#support-dl-bios and notice the "Improve memory compatibility" line, it means you can use higher RAM frequencies each time, now it doesn't tell much but I guess you would need the F3b or F3h BIOS for 6400Mhz.

Try updating the system/mainboard BIOS and then running it at setup defaults.

not a bug as sutch but a bad ram stick when being overclocked by XMP can cause BSODs and other sudden shut downs.

I have f1 version bios but why do you think to cause a problems because I see on list nothng mention about bug ram in version bios.
Any way I cahnge to 4800MHZ as base and will check for crushes on a game again and if it happned ill let you know.

Check cables on both ends (plugs/connectros/headers - if they are all are "locked", and secured in place).
Any places where you have tight bends ?
XMP enabled ?

Basicly, either PSUs overcurrent/overvoltage protection get's triggered (which switches off PC under load), OR your MB switches off PC because of some fault (maybe too high temp on VRMs ?).
I should add that motherboards with SPS style MOSFETs can be switch off by PWM controller itself (regardless of what VRM temps you see in diagostic tools, since not all inputs made by PWM Controler/MOSFETs can be checked in such tools).
Checking VRM's thermal pads may be required.

Lastly : RGB... overheating RGB specific controllers or "mixed" RGB/FAN/LED Strips ones, are also a thing to watch out for.

XMP enabled: true
i has 1000w psu and load game with 100% gpu and load paralel Cinebench so cpu was 100% too
and i had no crushes even two loads sequence multi t on Cinebench it can be dif sometimes if happend 5 min and sometimes it even 90 minutes (1 hour and half)

The temp cpu it low for regular case game load only my cpu 50-60C on game (that make crush - and it happened on dif games).

 

[RJARRRPCGP]

After SFC /scannow I found in the logs that a whole lot of files got corrupted (think it was net framework 4 files).
I had to do a (windows10) repair install to fix it , (using the keep settings and data option).

I suspect your SSD is faulty.

funny thing is yesterday i had a bsod after installing nvidias latest drivers, couldnt do anything the rigg would start get into windows and restart i had to go int safe mode option 5 and remove the drivers the install the older driver and alls good. nothing to relate to the OPs post but id thought id mention it just in case.

I suspect it's because you didn't block Windows Update from installing drivers. That's the first thing I do before getting the driver from the GPU maker's web site.

This is from TPU's own review, 90C is fine.

View attachment 280789

Mine's the hottest Vermeer, apparently, 82C is the same temp I usually got on Maxon Cinebench R23. Now, it looks like the 5900X is the better Vermeer, while I thought that they wouldn't be cooler, LOL.

 

[AsRock]

Were you running that memory at 6400 ?, that might be the reason good luck with 4800.

 

[Tom.699]

Does it happen in just one game you play or in many?
For me it does not seem to be a hardware problem.
May be just buggy game, memory leak or something.

 

[bug]

Generic advice: when the system misbehaves, set everything at default. Don't overclock anything (CPU, GPU, memory/XMP). Make sure it's stable. If it's stable, then you can set memory and the rest to OC, but one by one, confirming it's stable at each step.
If it's not stable at default settings, you're looking at a loose connector or some damaged hardware.