Post
Topic
Board Mining (Altcoins)
Re: PhoenixMiner 5.6d: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux)
by
relax69
on 18/08/2021, 07:00:47 UTC
Hi all,

I got a strange problem and am not sure if anyone here has the same issue.

My rig is running win10 with phoneix5.6d. Since I only mine 12 hrs a day to avoid peak time electric fees, it is auto turn on and off every day. The miner is autorun using the task scheduler after the rig is on. It has 10 gpus (6*rx580 + 4*gtx1070) so can generate around 300mhs. However, I have noted that when the rig is auto on and run the miner, it only got around 292mhs. I have to manually close the miner and run it again to get 300mhs, otherwise, it will stay around 289~292. I also found that the report hashrates in bitfly is unsteady when it's around 289~292. if I restart the miner (not rig), the report hashrates will be steadily staying at 300mhs      

For me, it sounds like the miner needs to be "warm up"?? anyone here got the same problem?? Btw, I remember I don't have this problem when using claymore.

For better answer you need to share your log files. I’m guessing from what you said possibly reason is, if you are using straps and rxboost they don’t running with automatic start until you start the rig manually with administrator rights.

this is my log file when running the miner the first time and got 292mhs
https://www.dropbox.com/s/d568k76xa8urlg1/log20210814_214923.txt?dl=0

this is when close and reopen the miner for a second time. I can get around 300mhs
https://www.dropbox.com/s/k2ej440ogvp4yjo/log20210814_224919.txt?dl=0

I don't use rxboost and strape. the setting is just a simple wallet address. I use afterburner for oc. I also noted that if the rig is running at 300mhs, which means the miner is restarted for a second time. The rig will become unstable, it will crash and reboot at least once 12hrs. after reboot it will back to around 292mhs unless close and open the miner again.



It looks like you have the Nvidia cards running the same in both examples but the difference is coming from the AMD cards. I don't like doing overclocking for mining with Afterburner, Watman, or third-party software because it makes the system unstable. PM is the best for me because it's a great tool for overclocking. I will recommend you do not use Afterburner at all or use it only for monitoring.
Also, recommend you to add this to your config file and give it try:


Code:
# RX 580, RX 580, GTX 1070, RX 580, GTX 1070, GTX 1070, RX 580, GTX 1070, RX 580, RX 580

 -powlim 0,0,75,0,75,75,0,75,0,0


  -tt 65
  -fcm 2
  -fanmin 50
  -tstop 88
  -tstart 65
  -hstats 2
  -hwm 1
  -fanstop 1
# -ttj 75
# -ttmem 79


 -mt 1,1,0,1,0,0,1,0,1,1
 -vmr 25
 -straps 1
 -vmt1 25
 -vmt2 20
 -vmt3 0

# -rxboost 1

 -cclock 1100,1100,-200,1100,-200,-200,1100,-200,1100,1100
 -cvddc 850,850,0,850,0,0,850,0,850,850
 -mclock 2075,2075,+700,2075,+700,+700,2075,+700,2075,2075
# -mvddc 750,750,0,750,0,0,750,0,750,750



I did not see your frequencies in the log files so, in the example, I used some information from the internet for the Nvidia cards because I never had GTX 1070 but you can replace the numbers from the Afterburner.
In the example for the RX 580, I use the settings I have for my Sapphire Pulse RX 580 with Samsung memory. Your settings may be different for different brands and memory, but you can replace clocks from the Afterburner.
I have my Sapphire Pulse RX 580 using only 83 Watts and running slightly above 30 Mh/s.

If these settings work for you well, you should lower your power usage too.