You got a couple things wrong. The machines are the same, from the september and november batch. They were just overclocked. I got one in sept and my friend got it a couple of days ago. It's hashing at 19.3 GH/s but at higher temps. In the miner status, what does the frequency tab say?
Also, the thing about error while reading temps...I'm guessing it's just a bug. I always get it if pool 0 dies and it has to switch, but it does not affect performance in any way. The hashboard stopping thing, I only got it 2-3 times.
I wouldn't mind if they were the same and overclocked as long as they were hashing at 19.3GH while still under warranty. The freq tab says "Default Setting" and the kernal log shows it running them at 487M, which if my math is right would be around 5750 per board?
It affects my performance when I get that temp read error, my hashrate drops to 11.6 GHs when one of the boards show errors, it's happening daily to at least one of the boards.
Your 3.5 issue I think tis the machine rebooting itself.
Have you tried downclocking?
What PSU are using? It is a solid 1600w PSU?
I figured they were rebooting, but when they ramp up, the hashrate never drops and I can keep "top" running in ssh so they don't seem to actually be rebooting.
The profits aren't great already, I can't imagine downclocking any more especially since they are running 1.8Ghs below expected.
I'm using two Bitmain AP3++'s on 220V and one 2400w HP Server PSU with breakout board on 220V
Your 3.5 issue I think tis the machine rebooting itself.
Have you tried downclocking?
What PSU are using? It is a solid 1600w PSU?
Not op, but I don't think so. Mine only does that if the pool dies somehow. We also need to know the conditions in which the machine is kept. What is the ambient temperature?
Correct, It doesn't seem to actually reboot. It does seem worse on Nicehash, but also does it running on antpool. Machines are in 65F (18c) basement, humidity steady at 35% with the dehumidifier