Well I don't know if it does make sense or help somebody but I have a workaround / solution. The Nitro does have a 4 Pin fan connection nearby the bios switch. If I connect a quiet Noctua fan to it it keeps spinning without noise and can be used to enforce continuously cooling to keep the temperatures down. Especially in the hot PWM area of the GPU. So it doesn't hysteresis anymore and keeps the card quite. Sure it will spin up under load but that's ok. The hysteresis was annoying.
Beiträge von hannibal1969
-
-
Wrong PWM value causes incorrect fan behavior which in turn causes overheating. Don‘t play with that stuff if you don‘t know what you are doing and use the card as is with stock settings.
I know what I’m doing. Pwm setting for nitro was mentioned to be 28 if i’m correct even told by you somewhere. I already tried all values in comment fields to be default. So everything to default. Even then it crashes under load. To be clear im happy in big sur without any device properties however the zero fan / turn on fan hysteresis is annoying. Maybe i should try it with other increased pwm value.
-
Darf ich fragen um welchen Riesenbeitrag es geht? Konnte den nicht finden...
Ich würde mich bei der Frage nach der Grafikkarte gerne anhängen. Zur Zeit habe ich eine Sapphire Nitro+ Vega 64 und bin recht unglücklich damit. Ich habe immer wieder das Problem mit den hochdrehenden Lüftern, Blackscreens und vorallem Renderfehler. ...
After putting this through google translate I have more or less the same issues. Although the Sapphire Nitro+ Vega 64 is a fantastic card in windows and nowadays usable in BigSur there are some annoying issues with it. Hopefully Chris can also give some tips. (please forgive my English instead of German because I'm Dutch ;-)),
1 As soon you use the PowerPlayTable (default) values and inject these in Device Properties the card will reset itself under high load. I think that use of Device properties and PowerPlay values changes somehow an unknown value which causes the card becoming too hot. All fans will go to 100% and display is gone but system itself is still active though.
2 When you don't use PowerPlayTable values the system is good usable but there is a hysteresis in warming and cooling the card. When you start the system the GPU is about 30 degrees. The temperature slowly increases to about 55/56 degrees C. Then the fans spin up a bit loud and everything's is 5 - 7 degrees cooled. And then this repeats itself over and over. You can understand that this is more annoying in summer compared to winter.
3 A flash of the 2nd bios of another card with standard cooler will avoid previous situation but can cause overheating because the temperature increases to a certain high level somewhere in the card which can crash the GPU.
4 With heavy load I mean a very intense game like Company of Heroes which can load the GPU to 100%. I don't know if many users in the past had tested the PowerPlay Table device properties under these kind of conditions. So maybe this occurs with everybody.
So indeed what are options to possibly solve this? I don't have any plans to order the too expensive new AMD GPUs.
Personally I think that a good refit of the cooler with decent thermal compound might already help...
