Question PC shuts down during gaming and RAM tests after upgrading from Ryzen 5 2400G to Ryzen 5 5600X ?

Jun 14, 2024
16
4
15
Recently ive been gradually replacing my components one by one ever since i bought a prebuilt which had a 1050ti, and a ryzen 5 2400g. first i bought a chieftec 700w psu, then a kingston 3200mhz 8 gig stick (there was a single adata 8gb 2400 mhz stick inside slready), a 2060 super and today i installed the 5 5600x and thats when the problems started ocurring.

At random times while playing demanding games, my pc just shuts off, like a power outage, no blue screen, no nothing. ive already ruled out most of what it could be, and i did stress tests on my cpu, gpu and psu which all came back with no errors.

i ran a memtest86, and there it was, about 10 seconds into the test, the pc shut off completely. to confirm my skepticism, i also did a memory test through occt and the same thing happened. While doing these tests with only one stick inside at a time, i had no errors, it goes smoothly. As soon as i put both sticks in, its a goner.

Setting an xmp profile in the bios doesnt even let the pc boot to the windows screen before it shuts off. the thing that confuses me the most is the way it shuts off, like it just loses power completely, the psu is relatively new, about a year old, and worked perfectly until i put the new cpu in. i was a bit inexperienced at the start when replacing components, hence the difference in ram sticks which i suspect to be the culprit. i could try and use my old psu which is rated for 550w, cause its something i still havent tried, but this whole ordeal smells like a ram issue to me.

regarding my previous inexperience, my kingston stick isnt even in the qvl list for my mobo (asrock a320m- dvs 3.0) but it was all still functioning till the day i replaced the cpu. im completely lost on what to do next, if someone could point me to the next step of troubleshooting, ill be eternally grateful.
 

boju

Titan
Ambassador
Shutdowns of this nature is more than likely power supply related and since that unit has 1yr warranty i am not surprised.

Where are you located? Should replace psu asap with something more reliable, even if it costs more. Psu is the most critical part, even if you don't see it, it is important to do it right and get a good one.

 
i bought a chieftec 700w psu
Chieftec has a lot of junk PSUs. Which model is it exactly?

then a kingston 3200mhz 8 gig stick (there was a single adata 8gb 2400 mhz stick inside slready
Mixing RAM is a no go with Ryzen. It causes instability and the 3200 MT/s is limited to the speed of the 2400 MT/s stick er both are even running slower.

asrock a320m- dvs 3.0
What BIOS version is the motherboard running?
 
  • Like
Reactions: Roland Of Gilead
Jun 14, 2024
16
4
15
I’m from Serbia. I mean i’ll bite the bullet and replace the PSU, but is there a way I can be a 100% sure that it’s causing the problem? You’re probably right considering the nature of the shutdowns, i’m just confused as to why doing a psu stress test for an hour straight gives no errors yet the pc just shits itself every time it’s something to do with memory (both on occt and the memtest86).
 
Jun 14, 2024
16
4
15
Chieftec has a lot of junk PSUs. Which model is it exactly?


Mixing RAM is a no go with Ryzen. It causes instability and the 3200 MT/s is limited to the speed of the 2400 MT/s stick er both are even running slower.


What BIOS version is the motherboard running?
It’s a chieftec GPC-700s.

Do you suggest i match the adata one since it’s in the qvl list of the mobo or get 2 new sticks altogether?

P7.40 ( The latest stable version of bios for this motherboard)
 

boju

Titan
Ambassador
I’m from Serbia. I mean i’ll bite the bullet and replace the PSU, but is there a way I can be a 100% sure that it’s causing the problem? You’re probably right considering the nature of the shutdowns, i’m just confused as to why doing a psu stress test for an hour straight gives no errors yet the pc just shits itself every time it’s something to do with memory (both on occt and the memtest86).

Possibly as power states change as worker load changes as in not constant, psu probably can't handle dips in power use or can't tolerate even at constant load. It's hard to say but if this is just cpu/ ram issue, not stable enough for such vigorous testing, it does happen, maybe not do that.

Still, 2 year warranty isn't comforting, it does speak of quality. 5 yrs should be minimum.
 
  • Like
Reactions: Roland Of Gilead
Jun 14, 2024
16
4
15
Possibly as power states change as worker load changes as in not constant, psu probably can't handle dips in power use or can't tolerate even at constant load. It's hard to say but if this is just cpu/ ram issue, not stable enough for such vigorous testing, it does happen, maybe not do that.

Still, 2 year warranty isn't comforting, it does speak of quality. 5 yrs should be minimum.
That’s what i thought at first honestly, because it only happens in scenarios like gaming, where the load frequently changes. I’m just surprised that it was working well with the 2400g up until i put the 5600x, very strange for sure. I might try the old PSU just to see if it changes the shutdown behavior, i’m surprised the old psu was working for a solid 6 years without stopping, considering it was also some trashy brand (rhino rx-550PA), but then again, i was running a 1050 ti at the time … My friend had the exact same issue and his solution was the PSU as well, but his pc had decent components from the start, not to mention his PSU was underpowered for the specs he was running. I’ll give it a shot, thank you
 
Jun 14, 2024
16
4
15
That's a horrible PSU. Very low quality group regulated unit with terrible voltage regulation and bad protections.


A new matched kit of 2.
I should’ve done more research instead of just looking at the wattage, that’s my fault lol. As for the ram, i’ll check the qvl and go from there, thank you
 

boju

Titan
Ambassador
Back in the Pentium 2 / 3 days I didn't take much notice of psus. These days you still have low power using parts and can get away with knockoffs. As you go up in more performance those reliances increase lol.
 
P7.40 ( The latest stable version of bios for this motherboard)
Did you clear CMOS after the bios update? Or when it was last updated? If not you will need to do that.

On the PSU, you should really get something quality. Any of Corsair TXM/RMx/RMi/RMe, Superflower Leadex 3, EVGA P5, Seasonic Focus GX, all of which have 7-10 year warranties. There are plenty of good quality units, that you will be able to bring to your next build.
 
Jun 14, 2024
16
4
15
Back in the Pentium 2 / 3 days I didn't take much notice of psus. These days you still have low power using parts and can get away with knockoffs. As you go up in more performance those reliances increase lol.
Yeah i guess nowadays you really have to be careful with parts, idek wtf i was thinking, shoulda jus built the whole thing from scratch instead of this bs 🤣
 
Jun 14, 2024
16
4
15
Did you clear CMOS after the bios update? Or when it was last updated? If not you will need to do that.

On the PSU, you should really get something quality. Any of Corsair TXM/RMx/RMi/RMe, Superflower Leadex 3, EVGA P5, Seasonic Focus GX, all of which have 7-10 year warranties. There are plenty of good quality units, that you will be able to bring to your next build.
probably the first thing i did after the bios flash, i did it again after the processor replacement too.

tbh if my 1050 ti and 5 2400g lasted this long, i’m gonna be sticking with this build for years to come, i just wanna make it as stable as possible, and if it means buying a good quality psu, then so be it i guess
 
  • Like
Reactions: Roland Of Gilead
if its shutdown caused by psu protection, then power button wouldnt work unless you flip psu switch
if powerbutton still works after shutdown, then mainboard protection kicked in..which can be caused by either power delivery or temperatures
That's a good point.

@ OP, I may have missed it. What mobo are you running? If the VRM/Power phases on the mobo aren't very robust, it could cause these issues too.
 
Last edited:
Jun 14, 2024
16
4
15
if its shutdown caused by psu protection, then power button wouldnt work unless you flip psu switch
if powerbutton still works after shutdown, then mainboard protection kicked in..which can be caused by either power delivery or temperatures
power button does work after shutdown, it doesn’t require the psu reset. i logged the temps with speedfan seconds before the shutdown (during gaming), and the gpu was at around 82c and the cpu around 54 ish, i’m not really sure, i’d have to check. afterburner limits the gpu temp to 83.
 
power button does work after shutdown, it doesn’t require the psu reset. i logged the temps with speedfan seconds before the shutdown (during gaming), and the gpu was at around 82c and the cpu around 54 ish, i’m not really sure, i’d have to check. afterburner limits the gpu temp to 83.
mainboard temperatures, there should be few, hwinfo should report those

gpu/cpu temperatures would just slowdown

also check 24pin and 8pin cpu connectors if inserted correctly and no burn marks there
 
Jun 14, 2024
16
4
15
The VRM's on that mobo are pretty weak. Only 4+2 power phases. Whilst I don't think that would prevent the CPU from working, I wouldn't even run PBO with it, as it could be causing your issue.
yeah i wasn’t planning on doing any oc-ing, is pbo enabled by default? cause i haven’t gone into the oc settings in the bios
 
yeah i wasn’t planning on doing any oc-ing, is pbo enabled by default? cause i haven’t gone into the oc settings in the bios
I can't be sure, as I don't know what config the bios was in before updates. But, It should be readily recognizable once you go through all the screen options in the bios. It will be under the OC Tweaker/Advanced tabs. It should be possible to turn it off it's on, or on if it's off.

https://download.asrock.com/Manual/A320M-DVS R3.0.pdf

Do a search in this PDF for Bios, and you will see all the screens and the explanations for each setting.
 
Jun 14, 2024
16
4
15
mainboard temperatures, there should be few, hwinfo should report those

gpu/cpu temperatures would just slowdown

also check 24pin and 8pin cpu connectors if inserted correctly and no burn marks there
might be a stupid question but, my ram stress test crashes my pc within 2-3 seconds, and other stress tests pass without issues, can the ram really put out that much power as to overheat the mainboard within seconds?