I’ve heard over time that K pros can get 15/T efficiency, but not much more info. If we can get something like 80T at 15 watts/T then you could get 5 S19 K pros for the same price as S21 XP, and get 400T for the same price.
If someone wants to do some testing, install Luxos which can be installed on any board using commander without SD, and make a custom profile and set both the voltage and frequency to its lowest. Then, keep the voltage on its lowest, but make a new profile and increase the frequency by 5. Check the chip health, if every chip is green then do the same again, add 5 frequency. Keep doing this until you get a yellow chip, which isn’t damage it just means the chip isn’t hashing at 100% at that voltage/frequency. So roll it back to the previous frequency that’s had 100% green chips, and read your wall meter and tell us the watts/T. Don’t rely on the power reader from the software. Also if you could tell us ambient temps that would be good too, and model (110t/115t/120t)
Hopefully someone does and shares results. This forum is good at helping people and right now I’m failing to see how us little guys can keep mining Bitcoin if you’re not paying an industrial rate. If we are doing the same shit as everyone else, especially at a worse electric rate then how will we ever get ahead?
Also the perks of this would be not only competitive efficiency but when price per PH is good you can run them on full speed and get 600T. Example at $70/PH at 8c/kw a S21 XP would pay $6.95 in power and return $18.9 and the k pros would cost $26.1 to run but return $42, so only a few extra $ at that price but the higher the price per PH the better it would be to run K pros on full power instead of S21 XPs. And daily break even at 8c on the XP is $26/PH and the K pros (assuming 15 watts/T) is $29/PH so a lot potential upside. Cheers