Upcoming Events

Weekly BMW races on Simracing.GP Weekly BMW races on Simracing.GP Other regular AC events on Simracing.GP Weekly GT3 Sprint Races on Simracing.GP Weekly GT4 Sprint Races on Simracing.GP

EU Mazda MX5 Cup @ Mosport - Wed 02 Dec 2020

Chris Down

Premium
Messages
2,550
Points
311
Very very similar problem, tried every thing to isolate, as you are.
Bit the bullet, bought a 1200w power supply, problem gone.
Yes I know, an overkill, but it was so frustrating, I decided to definitely remove the power supply from the list once and for all.
Also noticed how power hungry CPU and GPU’s were getting, that helped me make my mind up to.

Glad it sorted out your case. In general, the most important thing is the PSU quality over absolute wattage beyond a certain point -- I run a 10900K and 3080, both overclocked/overvolted -- which is more or less as much power as you're going to get on a non-SLI system --- on a 650W power supply without issue.

Lower quality power supplies often cannot supply power reliably as they approach their limits. I'd recommend Corsair's RM series, or the Seasonic Focus Plus Gold/Platinum.
 
Last edited:

Kek700

Premium
Messages
2,791
Points
302
I think the below is more a tale of tragedy, I think going mini itx was mainly the reason for the below, and using a mini itx case did not help?

I had to go full water cooling in an attempt to cool my CPU, ( custom pump etc). an overkill maybe, but even my Samsung 970 SSD was overheating, had to put a purpose-made fan to keep that cool.
Eventually, I ran an open computer case, so then it hardly came down to case cooling. Admittedly it was a particularly hot summer when I built the thing.
The main problem was thermal run away. Getting it to run at 70 to 75 deg C, but after a time with AC, especially with the hot climate at that time. I spotted it creeping up, then thermally running away to 90 deg C plus.
Because my then GPU a GTX1080 joined in and was noticeably downclocking. That is again why I went to an open case. Even the glass front of the open case had to be removed, it caused my GPU to run 5 deg C higher.
Why it was such a bleeding palaver? I felt forced to go down this route, the PSU sat at the end of this long drawn out road.
Now it seems reasonably reliable and runs at good sustainable temperatures for running AC and ACC. Even this summer.

My previous Computer with a gtx980ti and i5 6600k ran with virtually no problems at all in a closed case, Running triple monitors at over 100fps with ease. My 9600k and eventual rtx2080 proved to be a pain from the word go. All seems ok at the present time. :)

It would from my perspective, seem to be an exchange of performance for a power and temperature increase which seems to be the norm now.

The protracted waffle bares little resemblance to what was the drawn-out saga the above was.:(

I was running an expensive mini ITX 750W power supply that just could not except the maximum demands of the 9600K and 1080, or was it buggered.

How do you cope with an even more demanding 10900K and 3080 @Chris Down , especially the temperature demands, or was I just plain unlucky.:O_o:
 

Chris Down

Premium
Messages
2,550
Points
311
I was running an expensive mini ITX 750W power supply that just could not except the maximum demands of the 9600K and 1080, or was it buggered.

How do you cope with an even more demanding 10900K and 3080 @Chris Down , especially the temperature demands, or was I just plain unlucky.:O_o:

It sounds like the PSU was faulty, a 1080 + 9600K shouldn't come anywhere near 750W (by my calculations, max load in the worst case for those is less than 400W). Even if you add a bunch of fans and storage, I don't see how you're going to reach that.

As for temperature, a few things are important:
  • A decent case (so you can easily get wires out of the way which can reduce pressure loss a lot, and have decent fan placement)
  • Some decent fans (I mostly use Arctic P12 for radiators or other cases needing static pressure, and Arctic F12 for exhaust -- I still use P12 for intake since the airflow is not as noisy)
  • Fan placement: depends on your case, but I usually have 3 fans on the radiator (static pressure), 2 in on the bottom (intake airflow), and 2 on the top (exhaust airflow). Right now I have only 1 on the exhaust, but might add another when summer comes around again. F12/P12 is pretty much inaudible, so no concerns about adding more.

I also use an Arctic Freezer II 360, but it shouldn't be necessary if you don't OC the CPU.
 
Last edited:

Chris Down

Premium
Messages
2,550
Points
311
Wow! You prepping for a G2?

Heh, no plan to upgrade from my trusty Rift CV1 yet :) The foam is falling apart and all kinds of stuff, but I still love it. I guess I'm waiting for the next evolution of VR (probably the point where it can fill your vision completely), which hasn't happened yet.

I got the new computer mostly for Cyberpunk.
 
Last edited:

Chris Down

Premium
Messages
2,550
Points
311
It sounds like the PSU was faulty, a 1080 + 9600K shouldn't come anywhere near 750W (by my calculations, max load in the worst case for those is less than 400W). Even if you add a bunch of fans and storage, I don't see how you're going to reach that.

As for temperature, a few things are important:
  • A decent case (so you can easily get wires out of the way which can reduce pressure loss a lot, and have decent fan placement)
  • Some decent fans (I mostly use Arctic P12 for radiators or other cases needing static pressure, and Arctic F12 for intake/exhaust)
  • Fan placement: depends on your case, but I usually have 3 fans on the radiator (static pressure), 2 in on the bottom (intake airflow), and 2 on the top (exhaust airflow). Right now I have only 1 on the exhaust, but might add another when summer comes around again. F12/P12 is pretty much inaudible, so no concerns about adding more.

I also use an Arctic Freezer II 360, but it shouldn't be necessary if you don't OC the CPU.

Oh, it should be said I don't bother to run them that cool either. Typical Assetto Corsa values in a race are 72-73C on the 3080, and 95-96C on the 10900K, neither of which is a problem though.
 
Last edited:

GeekyDeaks

Premium
Messages
687
Points
677
I guess I'm waiting for the next evolution of VR (probably the point where it can fill your vision completely), which hasn't happened yet.
Yeah, I feel like I'm wearing a diving mask. It's not a major problem but it does mean I don't spot people to the side as easily as I would like. You get the odd comment from some that it's similar to wearing a helmet, but I can say categorically that it's not at all like it.

I got the new computer mostly for Cyberpunk.
Oh yeah, also eagerly awaiting that one!
 

GEO147

Premium
Messages
1,328
Points
301
@GEO147 , do you happen to know the make & model of your PSU. Have you had any problems with other games?

It's a Corsair AX860 about 5 years old transferred over from another build.
I've been looking into it and Fanalab was mentioned in the critical 41 error too.
 

GEO147

Premium
Messages
1,328
Points
301
So if I run Hardware Monitor during an AC session that will give me my temps?

Is there s record button or something or do I just leave the program open?
 

pattikins

Premium
Messages
1,295
Points
301
It's a Corsair AX860 about 5 years old transferred over from another build.
I've been looking into it and Fanalab was mentioned in the critical 41 error too.

It's a better PSU than mine (also manufactured by Seasonic) + it has a 7yr warranty & glowing reviews. It could be the PSU, even the best can develop issues, but I would be inclined to look elsewhere if the problem was isolated to playing AC. Have you looked in Windows' 'Event Viewer' to see if there are any recorded critical event errors that can be traced back to Fanalab (Fanatec wheel)?

I know you've got an AMD system, and like me, you've had some usb issues in the past. Have you tried plugging the wheel in to another USB (i.e., external powered usb hub) port?
 

GEO147

Premium
Messages
1,328
Points
301
Good points. Haven't changed anything as it's just happened the one time. But yeah there was a critical kernel power 41 in the event viewer.
And I recall seeing Fanalab mentioned.
 

pattikins

Premium
Messages
1,295
Points
301
Hopefully it was just a one off event. Have you done a search ('find') in event viewer for 'fanalab' to see if there other error/critical events that have gone unnoticed?
 
Top