My visible hashrate on D3 screen is much better with fewer hw errors using this firmware. However, my shares/poolside hashrate is better with stock firmware. I've ran 48 hour tests on the same pool using the same workers.
The only variable is the firmware, I wonder why this is so, because it seems very promising but for us not worth a 20% drop in shares.
This was tested using Blizz firmware previous version I have not had time to test this very latest. Anyone have insight on what could cause this?
I have experienced a reduction with some pools (mainly Yiimp clones) 5%. Switching seems to adversely lower their PPS calculations, but not 20%
No idea on the power usage, but I have kept this F\W as my D3's are running a lot cooler, no H\W errors ans no failed chains!!
Despite latest bitmain f\w, I was experiencing failed chains (all) 0 hash and fans on full!! I was running 487freq (batch 1)
Current settings;
Freq: 450, ASIC: Lowest, Fans: Auto
Stats:
M/H: 16.8ghz
H\W: Zero despite running for days
Temp (PCB): 59
Temp (Chip): 70
Fan: 3,690 (auto)
I can get the hash higher without H\W errors, but like how cool and stable the unit is running. 28 degrees ambient temps!!!
Perhaps as more users run the F\W, fees could reduce, but 1.5% seems reasonable for an excellent enhancement.
If you don't want the DEV fees, you can download the antminer toolkit from bitmain, to set the lower CHIP voltages, ect , but greater risk of getting it wrong and miss the key features, auto reboot on low hash, updated cgminer, better fan control, ability to set each board individually. Toolset can be a bit hard to get working with the D3's.