Topic: PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux) - page 478. (Read 784950 times)

Linux (ethOS) version request.
Thanks

I'm having an issue with 2.6. Basically when the watchdog detects a GPU thread that isn't responding the whole system locks up when the restart occurs. No issues on Claymore. Has anyone heard of this issue?

  Certainly voltages and/or clocks aren't the same. If there is any difference in the wattage it would be very small, nowhere near what you are seeing. Please check with GPU-Z, which seems to show the most accurate values of clocks/voltages.

  We haven't seen this on devfee but if the cards are paused even for a few seconds this may happen. During the switch between devfee and normal mining (and vice versa) it is possible to "hit" a small period of time and the cards to be paused for about 0.2-0.5 s, which may be causing what you are seeing. This is being redesigned in version 2.7 to completely eliminate the switching period, which should solve this problem as well as any instabilites that some have experienced on devfee periods even without DAG switching.

 Adding to the TODO list: Sleep less, work more, beat the competition

  Yep, we are seeing all kinds of this "funny" behavior with the Overdrive API when we test voltage control in 2.7. We have given all hope that there will be any driver where everything works without any weird stuff and random bugs, so we are including all kinds of workarounds in order to compensate for the buggy drivers.

 Our 1060s 3 GB works fine on ETC even with -eres 2. Please check if the DAGs are created properly if you exclude the card that is used as primary display adapter (i.e. the one that has a monitor or HDMI plug connected) - you can test the cards one by one using the -gpus option. Also check the answer to question P007 on the first post in this thread. Finally, go to Control Panel -> Advanced system settings -> Performance -> Visual Effects and turn off everything to lower the GPU memory usage by Windows itself.

 Do you mean the effective hashrate or the reported hashrate of the cards? And how long the rig is staying disconnected from the rig?

If miner disconnect from pool and re connec to pool hasrates drop half i have same problem 3 times in 2 days need to be fixed

My pool has constant difficulty of 10000MH

Hi, dev. Do you know about the problem with 1060-3 cards, that are not able to download dag file of ether classic? Would you solve the problem with reserving editional memory on videocard?

I just mine ETH + keccak. I was loosing like 1MH/s eth for 120MH/s Keccak in my cards, so it is not worth it for me. I haven't tried blake.

You are seeing that higher share rate afyer you start the miner or reboot because you are connecting to the pool which resets you the base difficulty for that port you have selected.  So say you mine for 10 hours you are on the 2G difficulty but over that last 10 hours the pool raised your difficulty to 4G because your rig(s) are faster, hense you solve less ahres but each solve is worth more heft.

Back to the restart and reboot, when the miner 1st starts up and you see that flurry of shares is because you are set by the pool at the lowest difficulty, either by the port you chose or all connect to one port and the pool auto adjusts.  regardless, every pool will auto adjust your difficulty.

So just because you get 100 shares in the first 30 minutes and 50 shares the next 2 hours its becasue your difficuty was increased.  You see this with phoenix when it submits a share, it will say the difficulty found and what your pool has you set at.  It will be low when you 1st connect and higher when you have been mining longer.  just remember as you go up in difficulty you solve less shares but the value of those higher difficulty shares are worth more, percentage wise in the PPLNS or whatever your pool is doing to share.

Reboot help me.
I don't know why. But if i first start claymore and close it and then start phoenix power usage very big.

Oh God, new Claymore with dual mining of blake and keccak and lower fee of 1.5% in dual. Guys I like your miner, but new Claymore with these algos is a game changer. Implement it and I will revert back to Phoenix. Probably we can see first fruits of concurency at this field. Thanks

Power consumption is exactly same like claymore.no difference at all.

GPUZ show 0.8875V
p.s. I also use a meter from the wall.

I get same power consumption from wall on both Phoenix and Claymore.  There should be something messed up with settings on yours.

p.s. I use an energy meter from the wall so I know the readings are accurate.

I use this params directly in bios. And when i use phoenixminer i run OverdriveNTool. All freq and voltage display like i wrote in bios. MSI AB give information like i wrote.

with your 1060 3gb hynix you can earn much more with mining other algo then ethash...

My rig have 6 rx588(samsung)
gpu freq 1200 mem 2100
voltage is 940 on gpu and mem

When i use claymore 10.6 eth+dcr
total hashrate : 183mhs on eth and ~5400 on dcr
1080-1100 watt from the wall

only ethereum mining

claymore 10.6
183-184mhs
850-870watt from the wall

phoenixminer 2.6
184-185 mhs
BUT power usage is
1200-1270 from the wall
Maybe this is realy fastest miner but it very expensive....

Can confirm, this was the fastest ETH miner I could find for my rig currently. (1060 3GB hynix)

Pretty good.

Getting 37mhash on a vega 56 WITHOUT blockchain drivers.

non uefi mobo so blockchain drivers don't work

I've noticed sometimes when I start it up the hashrate is stuck a bit low even after waiting a bit, seen 30-35 mh/s. When I restart it I get the full 39.7 mh/s. 1080 Ti.

Claymore has added keccak and blake2s to the dual mining list that are still not taken over by ASICs. Is there a possibility to add them to Phoenix as well?