Super Flower 1000W PSU - am I somehow rocking it too hard?

[GanjaSMK]

Long story short I snagged a 1000w Super Flower 80+Gold PSU back in 2021.  Packed up the PC went on the road.  Set it back up and attempting to full throttle the badboy. 

 

I noticed via HWiNFO64 that my GPU's reported amps are exceeding 105+, mostly spiking.  Seems to be also holding steady above 90A+ as in playing BF2042.  Randomly - not looking at amps when it happens of course, PC screens blackout and fans full up and have to hard reset.

 

Happening enough I'm really irritated and likely not going to play BF anymore.  That said, isn't happening in any other games.

 

So, paired with a 10850K at 1.3v, 6800XT and assuming the amperage readings are at least closely accurate - am I pushing the 83.5A rail too hard?

[UltraMega]

Try running Furmark and Prime95 at the same time and watch Furmark to make sure your GPU is staying at 99/100% usage while your PC is under heavy load. 

 

Could be something else though, maybe the SSD you have the game installed on is failing, or maybe it's just a driver issue. Any other symptoms? 

[ENTERPRISE]

13 hours ago, GanjaSMK said:

Long story short I snagged a 1000w Super Flower 80+Gold PSU back in 2021.  Packed up the PC went on the road.  Set it back up and attempting to full throttle the badboy. 

 

I noticed via HWiNFO64 that my GPU's reported amps are exceeding 105+, mostly spiking.  Seems to be also holding steady above 90A+ as in playing BF2042.  Randomly - not looking at amps when it happens of course, PC screens blackout and fans full up and have to hard reset.

 

Happening enough I'm really irritated and likely not going to play BF anymore.  That said, isn't happening in any other games.

 

So, paired with a 10850K at 1.3v, 6800XT and assuming the amperage readings are at least closely accurate - am I pushing the 83.5A rail too hard?

 

To ensure that it is not a windows/driver related issue, I would disable auto restart upon a blue screen. By default windows is set to auto restart when it hits an issue causing a BSOD, which I think is fairly idiotic setting as it makes the issue harder to diagnose. 

 

Disable auto restart by going to Control Panel > System > Advanced System Settings. Go to the "Startup & Recovery" > Settings and un-tick the "Automatically Restart" checkbox and click OK. 

 

Next time the PC reboots itself, a BSOD will appear and will indicate what driver/software failed and caused the BSOD. Hopefully this will give you some more clues.  If it happens, take a picture and pop it in this thread

 

IF it is a PSU issue then the system will trip and auto reboot, you would not get a BSOD in this instance. 

[GanjaSMK]

Thing is there is no BSOD - it's like the graphics card just kills the system; monitors go black, audio persists, then stops, system halts and fans rev up and i have to hard reset.

 

@UltraMega  I'll do that and see what's happening.

 

Also, I realize after doing a little reading on general electricity, if I was pulling 100Amps I'd be vaporized in explosions.  So I'm wondering if the reading in HWiNFO64 isn't actually AMPS or maybe even is, but isn't reporting correctly, due to some random way the app/MB sends/processes signals.

Edited June 25, 2023 by GanjaSMK

[EHW Ai]

This sounds like an issue with the GPU not the PSU.

 

What is event viewer showing?

[pio]

My 6900XT seems to do the same thing on occasion.  I have yet to determine an actual fault.  In my case, a reboot usually fixes it.

[UltraMega]

This is the kind of issue that could really be anything. If it's a hardware issue, it's definitely more likely to be the PSU or GPU, but could be ram, mobo, whatever. The best thing you can do for now is to try to replicate the issue is some other way or find any other evidence of the issue at all to help narrow it down. 

 

Remind me of a buddy of mine who had a PC that kept crashing in PUBG and we tried everything to resolve it to the point of replacing just about everything except the SSDs and it turned out to be an SSD that was causing the issue. Doubt that's the case here, just reminds me of that. 

[ENTERPRISE]

8 hours ago, GanjaSMK said:

Thing is there is no BSOD - it's like the graphics card just kills the system; monitors go black, audio persists, then stops, system halts and fans rev up and i have to hard reset.

 

@UltraMega  I'll do that and see what's happening.

 

Also, I realize after doing a little reading on general electricity, if I was pulling 100Amps I'd be vaporized in explosions.  So I'm wondering if the reading in HWiNFO64 isn't actually AMPS or maybe even is, but isn't reporting correctly, due to some random way the app/MB sends/processes signals.

Hmm seems like it will be a process of elimination then. Are you able to get your hands on another PSU for testing? I think if you can eliminate the PSU as the issue that's a good head start.  

 

Also, do you experience these issues running everything at Stock ? No OC or XMP set ? 

[GanjaSMK]

3 hours ago, UltraMega said:

This is the kind of issue that could really be anything. If it's a hardware issue, it's definitely more likely to be the PSU or GPU, but could be ram, mobo, whatever. The best thing you can do for now is to try to replicate the issue is some other way or find any other evidence of the issue at all to help narrow it down. 

 

Remind me of a buddy of mine who had a PC that kept crashing in PUBG and we tried everything to resolve it to the point of replacing just about everything except the SSDs and it turned out to be an SSD that was causing the issue. Doubt that's the case here, just reminds me of that. 

 

Yeah super annoying.  I'm hoping RAM; that said could be mobo, GPU, etc. I actually have a new kit otw and gonna see if that proves a more robust/stable system. SSD sensors/reads all show no issues, no corruption via SFC/chkdsk/DISM; and temps are all in line.  Bummer no less.

 

33 minutes ago, ENTERPRISE said:

Hmm seems like it will be a process of elimination then. Are you able to get your hands on another PSU for testing? I think if you can eliminate the PSU as the issue that's a good head start.  

 

Also, do you experience these issues running everything at Stock ? No OC or XMP set ? 

 

Yup, happens at stock, XMP/no XMP, various voltage configs; no crazy temps above 90c max, I don't get it.

 

After previous post I ran DDU and reinstalled drivers, kinda tempted to try BF2042 again but also... not super stoked to have it go awry if it wasn't some random driver thing and is hardware related.

Edited June 26, 2023 by GanjaSMK

[EHW Ai]

IIR BF really pushes ram to the max, does your ram pass TM5 anta extreme or absolut?

[GanjaSMK]

Well, new RAM kit came and sadly the Micron chips don't have much wiggle room at all; stock CL18 @ 4600 wants my SA/IO voltage waaaay too high since DR dimms on this particular MOBO scale (DR, 16GBx2) top out at 4400. 

 

And since my B-DIE kit is capable at 4200/CL16 without error... I guess I'm now selling the new kit I got!  LOL..

 

Lastly I ended up uninstalling BF2042.  I tried so many things... limiting core/thread counts from 20/16 to 20/8, still crash hard.  Tried overclocked, stock, stock RAM and even with the 10850K and 2933.... it. still. crashes. 

So.... bummer I can't play BF2042 as I do enjoy the game.  Crashing no fun though.  @UltraMega You mentioned a buddy had a similar instance with a different game and the SSD ended up being the culprit.  Got any more info on what exactly the SSD was doing causing the issue or was it just *bad data*?

[UltraMega]

2 hours ago, GanjaSMK said:

Well, new RAM kit came and sadly the Micron chips don't have much wiggle room at all; stock CL18 @ 4600 wants my SA/IO voltage waaaay too high since DR dimms on this particular MOBO scale (DR, 16GBx2) top out at 4400. 

 

And since my B-DIE kit is capable at 4200/CL16 without error... I guess I'm now selling the new kit I got!  LOL..

 

Lastly I ended up uninstalling BF2042.  I tried so many things... limiting core/thread counts from 20/16 to 20/8, still crash hard.  Tried overclocked, stock, stock RAM and even with the 10850K and 2933.... it. still. crashes. 

So.... bummer I can't play BF2042 as I do enjoy the game.  Crashing no fun though.  @UltraMega You mentioned a buddy had a similar instance with a different game and the SSD ended up being the culprit.  Got any more info on what exactly the SSD was doing causing the issue or was it just *bad data*?

IDK what the SSD was doing exactly, as far as I know there were no other obvious symptoms. Maybe it had a bad sector that its own self checking didn't detect. 

 

Did you try running prime95 and Furmark at the same time? Does the GPU stay at 100% usage? 

Edited June 29, 2023 by UltraMega

[GanjaSMK]

Yep, stays pegged no problems. I just don't get it... lmao.  Oh well... no BF2042 for me.

[UltraMega]

1 hour ago, GanjaSMK said:

Yep, stays pegged no problems. I just don't get it... lmao.  Oh well... no BF2042 for me.

Are you able to try a different ssd? 

[GanjaSMK]

At this moment, I could swap one from my laptop, or buy another; but I'm headed out of town for a few days and thinking about buying one after I'm back.  Though, I should note I did install (reinstall, no copy/move) on different SSD's (3) to see if it was specific to one... but same results.

[UltraMega]

1 minute ago, GanjaSMK said:

At this moment, I could swap one from my laptop, or buy another; but I'm headed out of town for a few days and thinking about buying one after I'm back.  Though, I should note I did install (reinstall, no copy/move) on different SSD's (3) to see if it was specific to one... but same results.

Probably not an ssd tbh but if you can test two different drives without having more than one plugged in at a time you can definitively rule that out. 

 

What else have you tried? 

[GanjaSMK]

Pretty much everything.... Voltage, auto, GPU downclocks (along with stock/OC settings), pretty much the works so far.  No dice.  SA/IO voltages... all of it.
 

Something wonky.  Dunno what or how to figure out at this point. That's why uninstall.

 

[UltraMega]

6 hours ago, GanjaSMK said:

Pretty much everything.... Voltage, auto, GPU downclocks (along with stock/OC settings), pretty much the works so far.  No dice.  SA/IO voltages... all of it.
 

Something wonky.  Dunno what or how to figure out at this point. That's why uninstall.

 

Did you try another GPU or PSU? Did you try one ram stick at a time? Bios update? 

[GanjaSMK]

1 hour ago, UltraMega said:

Did you try another GPU or PSU? Did you try one ram stick at a time? Bios update? 

 

BIOS is up to date (most current), RAM is all good.  PSU good I think the issue could possibly be related to MPO (suggested to disable via registry) - I just found this via another thread. Seeing if it fixes/helps. Hoping it does!

 

Downloading...

[GanjaSMK]

Well... there goes my old brain. Did the one thing I hadn't tried... reverting to older drivers.  Went back two I think and seems to resolve the issue.

 

*fingers crossed*

 

Also disabling MPO did nothing for my case, result still same on most current driver 23.5.2. 

 

 

[UltraMega]

Nice, good job. Drivers are so stable these days, it hardly ever occurs to me to try older ones anymore. 

[ENTERPRISE]

Glad you got it sorted, though usually if its a driver issue, Windows "should" BSOD, so im surprised the driver issue tripped the system into a full reboot automatically. That was why I bought up making sure that auto reboot was disabled. 

 

Strange, but all that counts is you got it sorted and you can play BF !

[GanjaSMK]

Stoked the driver fixed it because... i do love wasting hours killing people in BF lol.

[UltraMega]

I'd play some BF with you. I haven't been into it for months, the rail gun update killed it for me because it became super unbalanced at the time. Hopefully it's better now. 

[GanjaSMK]

Right on I'm down I'll hit you up