Topic: PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux) - page 377. (Read 784958 times)

The option for lower power consumption is -clgreen 1
It is not clkernel -1

It seems -clkernel 1 option is not working for my rx570/580 4gb. The power consumption on-the-wall is still the same. I also didnt put -gt option in my config.txt

UPDATE:
Try -clkernel 2 and auto-tune gpu again.
And still the same hashrate and power consumption.

Can you add MOAC and CLO coins please? Thanks.

I think it's because there is a -tt 0 in there...

The top part of the config is a carry over from a global_config.txt used with claymore. I've edited my global_config.txt to only have:

I had the same problem with the webgui not showing fans and temps. At first I had -tt 0 in the user config because I thought I needed it using HiveOS autofan. Using -tt 0 obivuosly won't output any temp/fan stats.

Using -tt 1 in phoenix or not specifying any -tt (guessing default is -tt 1) just like claymore  , it displays temps and fans fine.

Discussion over at HiveOS: https://forum.hiveos.farm/t/custom-miner-integration/4521/257

-clkernel 2 = 13.157 MH/s
-clkernel 1 = 10.856 MH/s
-clkernel 0 = 8.858 MH/s

Not bad at all, but still lower than Claymore.


I'll be glad to help. Let me know if you need more logs, test new build or something else.

   The config seems OK apart from the two -mport options but the second one just overrides the value of the first one, so this isn't really a problem. As for the missing temperature and fan data - this is quite strange, we haven't seen this - perhaps this depends on how hiveOS extracts these: from the miner, or independently. It seems that hiveOS 0.5-74 is based on Ubuntu 18.04, which may be cause or at least part of the problem - we are testing now with Ubuntu 18.04 LTS, so we hope that we will find and fix any problems.

   As for the -tt not working - this is normal, as we do not support hardware control under Linux yet. It will be added in the next version.



   Unfortunately, the two RX550 that we have here are both with the Baffin core, not the gfx804, so we can't test this directly. You can try to use the alternative kernels (-clkernel 2) to see if this will solve the problem. We will try to source a gfx804 based RX550 to see what is going on.




   It is possible but we need more information to check and solve the problem. Please let us know the Linux version, the version of the driver, and what cards you are using.




   If you are using Linux, this is normal, as we don't support -tt at all in this version - it will be added in the next version. However, if this is under Windows, and you are using older driver (anything older that 18.x.x), you should upgrade to 18.x.x driver because the HW control support before that is broken in the drivers.




   The config seems fine. You can test the stability of DAG creation by running PhoenixMiner.exe -bench 2, and then every time when you press the d key in the PhoenixMiner window, the DAG epoch will increase. If you are able to go through several DAG increases without crashing, the the configuration is stable enough.



   Dual mining is the next big feature in our list. We are fairly confident that at least the first two algorithms will be ready within a month.


 

Depends how many cards you have. If you have 6+ then yes you should increase it. The page file has to be big enough to hold the DAG for all GPUs in the system.

Current DAG size is 2.65 GB

6 cards = 15.9 GB = 16281 MB
7 cards = 18.6 GB = 18995 MB
8 cards = 21.2 GB = 21709 MB

And so on those numbers will only get larger as the DAG continues to increase.

I'm trying this, waiting for the next epoch  thank you 

My page file set at 20000MB, do I need to increase? 

when can we dual mining? hurry up and you will just launch this dual version.
I'm tired of using claymore.

Already done. OC 1200/2050 in both miners. The card's flashed by "One click" in Polaris.

Also try increasing the size of your page file.

try:
- eres 0
- lidag 3

Here is my Phoenix miner config   my rig running XFX RX470 get error recreate DAG when epoch switching   How to fix this  

@PhoenixMiner
look at the task manager windows
PhoenixMiner 3.0c why gpu copy %100 ?
Claymore 11.9 - gpu copy %0

You have to find external OC settings tool for Linux like msi afterburner for windows.

It seems to me that OC settings doesn't work yet under Linux. Under Windows they work fine and PM is faster or consumes less energy than Claymore.

Asus Strix Radeon RX580 8GB Gaming, this cards works incorrect with -tt parameter, if i set -tt 60, fan speed jump from 0% to 41% only, and cards are trying cool yourself to 51-54 degrees   Same situation with Vega 64 Asus Strix.
XFX Radeon RX 580 GTR 8GB, this cards are working correct with -tt 60, but if temperature increases on cards, fan speed stopped on 59% and the cards begin to raise the temperature higher the limit 60. Same situation with other XFX cards.

I'm on Linux and I just noticed Phoenix is slower than Claymore

Hope it's only because Linux version is new

Any advice? mining on NVIDIA

Wait for what? Auto-tuning is over at 2018.09.13:14:54:14.570. Hashrate will not encrease any more.


OC settings are the same.

also adjust cclock mclock and cvddc parameters to be the same as in claymore's configuration.