PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux)

[bategojko74]

Hello, I'm testing Phoenix with a 8 mixed AMD RX570/RX580 rig, coming from Claymore. Everything seems good and little bit faster BUT I have one problem when starting the miner: unless I disable the watchdog timer with -wdog 0 , everytime I have the message that one or two GPUs are not responding and the miner resets while creating DAGs. Is there a way to make the wd wait longer or another fix?
Thank you

Try -gser 3 or higher value less than 11

Sorry but it didn't work, always crashing with watchdog activated, anyone? I can't believe it's only me having this issue

Have you tried -gser 0 or 1 or 2

[bategojko74]

Hello, I'm testing Phoenix with a 8 mixed AMD RX570/RX580 rig, coming from Claymore. Everything seems good and little bit faster BUT I have one problem when starting the miner: unless I disable the watchdog timer with -wdog 0 , everytime I have the message that one or two GPUs are not responding and the miner resets while creating DAGs. Is there a way to make the wd wait longer or another fix?
Thank you

Try -gser 3 or higher value less than 11

Sorry but it didn't work, always crashing with watchdog activated, anyone? I can't believe it's only me having this issue

run two instances of phoenix miner. First instance must have setting -gpus 1234 and all other settings the same as your current. Second instance must have setting -gpus 5678 and all other the same. If you use config file, copy it and in the first config1.txt put -gpus 1234 in the second config2.txt put -gpus 5678, after that run first instance by: PhoenixMiner.exe config1.txt and second instance by: PhoenixMiner.exe config2.txt

[dhmctrader]

Hello, I'm testing Phoenix with a 8 mixed AMD RX570/RX580 rig, coming from Claymore. Everything seems good and little bit faster BUT I have one problem when starting the miner: unless I disable the watchdog timer with -wdog 0 , everytime I have the message that one or two GPUs are not responding and the miner resets while creating DAGs. Is there a way to make the wd wait longer or another fix?
Thank you

Try -gser 3 or higher value less than 11

Sorry but it didn't work, always crashing with watchdog activated, anyone? I can't believe it's only me having this issue

Your not the only one...

Trying to switch back to ETH from another coin... only with -wdog 0 could i get running on ETH.

[PhoenixMiner]

The first beta of the 4.1 is ready. You can download PhoenixMiner 4.1a from here:

(MEGA links are no longer active)

If you want to check the integrity of the downloaded file, please use the following hashes:

Code:

   File: PhoenixMiner_4.1a_Windows.zip
    ===================================
   SHA-1: 4ffa1fb84c351cd040386120f1b40f1724fdb3ab
 SHA-256: 9940a8644293a73aa2d4f75152a3a5fde95610d2d168bcf882aff6f06af4beeb
 SHA-512: fc34724cf4b841c145f04e0fdc59df6261fa715466ccdb1f430a2ca7584b7ab2da0e24b3832815f813dbbc27dc5c517b3c315897d420c2d0644f02b54236ee45

    File: PhoenixMiner_4.1a_Linux.tar.gz
    ====================================
   SHA-1: e5d265ee97000e4e231d8c5af3ebbc2d5abeb9bc
 SHA-256: 8875882c2ff7714eafd26c11aa68115ff453159b2a3ca8464469dd3c1f7c3b22
 SHA-512: f2527a598bb2875c127331a785c91907dc2b3eadd1115308c0ad42f762f9c8adc06b87934f0488fb337354b47b43df60c630478555508590182c4916d87d237e

Here are the release notes for this release:

• Added faster "turbo" kernels (-clkernel 3) for AMD 470/480/570/580 GPUs. These are usually faster than normal kernels (-clkernel 1 or -clkernel 2) but require double the amount of VRAM. This means that you can only use them on 8 GB cards if mining ETH or ETC. You can use them on cards with 4 GB or less VRAM only if you are mining low DAG epoch coins.
• Added Blake2s dual mining kernels, and green kernels for AMD Fury GPUs
• Added Blake2s dual mining kernels for AMD 280/285/290/380/390 GPUs
• New faster kernels for AMD 280/290/390 GPUs
• Added alternative kernels (-clkernel 2) for AMD 280 GPUs
• Added options -pauseat and -resumeat for scheduled pausing and resuming of the miner. Example: -resumeat 22:00 -pauseat 6:00 - the miner will mine only between 22:00 and 6:00. Example 2: -resumeat 22:00,12:00 -pauseat 6:00,16:00 - you can specify multiple pause and resume times.
• (Windows only) The miner now prevents the computer from sleeping if the mining is not paused, and wakes the computer when it must resume mining at the time specified with -resumeat command-line option. In combination with -pauseat and -resumeat options this allows you to mine only when the electricity rates are lower and to keep your rig(s) asleep the rest of the time.
• When the miner is paused for more than a few minutes, the pool is disconnected. The pool connection is restored when the mining is resumed.
• Added a watchdog protection against freezing of the network connection thread
• Automatic selection of compatible devfee coin if there isn't enough VRAM to mine ETH and the main coin can't be recognized or isn't supported for devfee
• Automatic switch to the next pool if the cards VRAM is not enough for the requited DAG epoch. This is especially useful if the pool is auto-switching the coins but your cards only support lower DAG epoch coins.
• Added a -rvram option to set the minimal reserved VRAM that can't be used for mining. The default values are 384 MB on Windows, and 128 MB on Linux. You can also set -rvram -1 if you want to turn off the protection against using too much VRAM.
• Added support for Windows AMD drivers 18.12.x
• Other small improvements and fixes

[bategojko74]

The first beta of the 4.1 is ready. You can download PhoenixMiner 4.1a from here:

• Added faster "turbo" kernels (-clkernel 3) for AMD 470/480/570/580 GPUs. These are usually faster than normal kernels (-clkernel 1 or -clkernel 2) but require double the amount of VRAM. This means that you can only use them on 8 GB cards if mining ETH or ETC. You can use them on cards with 4 GB or less VRAM only if you are mining low DAG epoch coins.

Thanks
plus ~200 hesh from a video card with autotuning enabled

200 hesh is nothing. Maybe you mean 0.2 MH = 200 KH. I have 0.2 MH increase on RX570 8GB with PM 4.1a.

[Guigs321]

Same here 0,3 Mh/s earned per GPU (RX580 8 Go) !

[plastkort]

will you have support for ProgPoW?

[Ruvv]

Need to enable the user to increase the check interval for -wdog, for example -wdogtime 90. Now it costs 30 seconds, sometimes it’s not enough

[PhoenixMiner]

will you have support for ProgPoW?

  Yes, absolutely. We will be ready for the future switch to ProgPow.

Need to enable the user to increase the check interval for -wdog, for example -wdogtime 90. Now it costs 30 seconds, sometimes it’s not enough

  We will add a similar option in 4.1b, which will be released in a few days.

It should be noted that in this mode -clkernel 3, with the default setting -rvram video cards consume more power on average 2-4% and there will be more virtual memory loading, so maybe someone will need to increase it

   Yes, depending on your overclocking settings the turbo kernels may consume more power. Also, you must re-run auto-tuning because the optimal -gt values will most likely be quite different for the new kernels.

Note that you must also re-run auto-tune if you have AMD 280/285/290/380/390 GPU, because of updated kernels for these GPUs.

[big_daddy]

are you getting better results with the blockchain drivers or with the lastest AMD drivers? RX4xx & RX5xx cards

[UselessGuru]

 We will add a similar option in 4.1b, which will be released in a few days.

Could you please add the resulting autotuning parameters to the API?
Thaaaaaaaaank you!

[gsrcrxsi314]

Still no overclock or voltage control on nvidia cards?

Ethos has been able to do this for a long time.

[UselessGuru]

 We will add a similar option in 4.1b, which will be released in a few days.

Please suppress the warning messages about invalid epools.txt when valid epool parameters are passed via command line.

Eth: Invalid wallet 'YourWallet.BLACKBOX' specified for pool 1 from epools.txt. Replace with a real wallet
Eth: Invalid wallet 'YourWallet.BLACKBOX' specified for pool 2 from epools.txt. Replace with a real wallet

Code:

# The pools specified in this file will be added to the pools specified with the PhoenixMiner's
# command-line options (see -pool and -pool2 command line options).

When -epool -or -dpool is present, then no other pools should be used.

Thank you!

[UselessGuru]

Typo:

'commnad' should be 'command'

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
No avaiable GPUs for mining. Please check the -gpus commnad line option.

[UselessGuru]

turbo kernel error clCreateKernel (-46):

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
^C
c:\Users\Stephan\Desktop\MultiPoolMiner-Test\Bin\AMD_NVIDIA-PhoenixminerEthash_v4.1a>PhoenixMiner.exe -mport -4004 -proto 1 -pool europe.ethash-hub.miningpoolhub.com:20565 -wal ***************** -pass x -dcoin blake2s -dpool blake2s.eu.nicehash.com:3361 -dwal *************************.multipoolminer -dpass x -sci 60 -gt 0 -log 0 -wdog 0 -clkernel 3 -gpus 3
Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
...
Light cache generated in 0.9 s (26.0 MB/s)
GPU1: Allocating DAG (2x 1.44) GB; good for epoch up to #56
GPU1: Generating DAG for epoch #54
GPU1: DAG  50%
GPU1: DAG 100%
GPU1: DAG generated in 3.2 s (452.0 MB/s)
GPU1: Unable to prepare kernels: clCreateKernel (-46)

Card is a RX 570 4 GB model. Should this not be enough memory?

[bategojko74]

turbo kernel error clCreateKernel (-46):

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
^C
c:\Users\Stephan\Desktop\MultiPoolMiner-Test\Bin\AMD_NVIDIA-PhoenixminerEthash_v4.1a>PhoenixMiner.exe -mport -4004 -proto 1 -pool europe.ethash-hub.miningpoolhub.com:20565 -wal ***************** -pass x -dcoin blake2s -dpool blake2s.eu.nicehash.com:3361 -dwal *************************.multipoolminer -dpass x -sci 60 -gt 0 -log 0 -wdog 0 -clkernel 3 -gpus 3
Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
...
Light cache generated in 0.9 s (26.0 MB/s)
GPU1: Allocating DAG (2x 1.44) GB; good for epoch up to #56
GPU1: Generating DAG for epoch #54
GPU1: DAG  50%
GPU1: DAG 100%
GPU1: DAG generated in 3.2 s (452.0 MB/s)
GPU1: Unable to prepare kernels: clCreateKernel (-46)

Card is a RX 570 4 GB model. Should this not be enough memory?

Dual mining doesn't support clkernel 3

[UselessGuru]

turbo kernel error clCreateKernel (-46):

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
^C
c:\Users\Stephan\Desktop\MultiPoolMiner-Test\Bin\AMD_NVIDIA-PhoenixminerEthash_v4.1a>PhoenixMiner.exe -mport -4004 -proto 1 -pool europe.ethash-hub.miningpoolhub.com:20565 -wal ***************** -pass x -dcoin blake2s -dpool blake2s.eu.nicehash.com:3361 -dwal *************************.multipoolminer -dpass x -sci 60 -gt 0 -log 0 -wdog 0 -clkernel 3 -gpus 3
Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
...
Light cache generated in 0.9 s (26.0 MB/s)
GPU1: Allocating DAG (2x 1.44) GB; good for epoch up to #56
GPU1: Generating DAG for epoch #54
GPU1: DAG  50%
GPU1: DAG 100%
GPU1: DAG generated in 3.2 s (452.0 MB/s)
GPU1: Unable to prepare kernels: clCreateKernel (-46)

Card is a RX 570 4 GB model. Should this not be enough memory?

Dual mining doesn't support clkernel 3

Thank you! Is this missing in the documentation, or did I just miss it?

[bategojko74]

turbo kernel error clCreateKernel (-46):

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
^C
c:\Users\Stephan\Desktop\MultiPoolMiner-Test\Bin\AMD_NVIDIA-PhoenixminerEthash_v4.1a>PhoenixMiner.exe -mport -4004 -proto 1 -pool europe.ethash-hub.miningpoolhub.com:20565 -wal ***************** -pass x -dcoin blake2s -dpool blake2s.eu.nicehash.com:3361 -dwal *************************.multipoolminer -dpass x -sci 60 -gt 0 -log 0 -wdog 0 -clkernel 3 -gpus 3
Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
...
Light cache generated in 0.9 s (26.0 MB/s)
GPU1: Allocating DAG (2x 1.44) GB; good for epoch up to #56
GPU1: Generating DAG for epoch #54
GPU1: DAG  50%
GPU1: DAG 100%
GPU1: DAG generated in 3.2 s (452.0 MB/s)
GPU1: Unable to prepare kernels: clCreateKernel (-46)

Card is a RX 570 4 GB model. Should this not be enough memory?

Dual mining doesn't support clkernel 3

Thank you! Is this missing in the documentation, or did I just miss it?

I found it experimentally. Looks like bug.

[lesjokolat]

will you have support for ProgPoW?

  Yes, absolutely. We will be ready for the future switch to ProgPow.

Need to enable the user to increase the check interval for -wdog, for example -wdogtime 90. Now it costs 30 seconds, sometimes it’s not enough

  We will add a similar option in 4.1b, which will be released in a few days.

It should be noted that in this mode -clkernel 3, with the default setting -rvram video cards consume more power on average 2-4% and there will be more virtual memory loading, so maybe someone will need to increase it

   Yes, depending on your overclocking settings the turbo kernels may consume more power. Also, you must re-run auto-tuning because the optimal -gt values will most likely be quite different for the new kernels.

Note that you must also re-run auto-tune if you have AMD 280/285/290/380/390 GPU, because of updated kernels for these GPUs.

any plans for kernal 3 for fury x or vega 64?

also the setting of etth (claymore lower limti of 50 being lowered to 25?) i think can help squeeze out some more hashs..

thanks

[PhoenixMiner]

Could you please add the resulting autotuning parameters to the API?
Thaaaaaaaaank you!

   There will be substantial additions to the remote management API but not in this release.

Still no overclock or voltage control on nvidia cards?

Ethos has been able to do this for a long time.

    We are giving priority to any ideas to bring up the ROI of the mining but these will be implemented eventually.

Please suppress the warning messages about invalid epools.txt when valid epool parameters are passed via command line.

Eth: Invalid wallet 'YourWallet.BLACKBOX' specified for pool 1 from epools.txt. Replace with a real wallet
Eth: Invalid wallet 'YourWallet.BLACKBOX' specified for pool 2 from epools.txt. Replace with a real wallet

Code:

# The pools specified in this file will be added to the pools specified with the PhoenixMiner's
# command-line options (see -pool and -pool2 command line options).

When -epool -or -dpool is present, then no other pools should be used.

    This would change the way the miner behaved in all current releases, which will definitely cause problems for many users. You can either rename the epools.txt to something else, or comment out all lines there with the # character at the start of each line.

turbo kernel error clCreateKernel (-46):

Code:

Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
^C
c:\Users\Stephan\Desktop\MultiPoolMiner-Test\Bin\AMD_NVIDIA-PhoenixminerEthash_v4.1a>PhoenixMiner.exe -mport -4004 -proto 1 -pool europe.ethash-hub.miningpoolhub.com:20565 -wal ***************** -pass x -dcoin blake2s -dpool blake2s.eu.nicehash.com:3361 -dwal *************************.multipoolminer -dpass x -sci 60 -gt 0 -log 0 -wdog 0 -clkernel 3 -gpus 3
Phoenix Miner 4.1a Windows/msvc - Release build
-----------------------------------------------

CUDA version: 10.0, CUDA runtime: 8.0
OpenCL driver version: 18.12.2,
Available GPUs for mining:
GPU1: Radeon RX 570 Series (pcie 8), OpenCL 2.0, 4 GB VRAM, 32 CUs
...
Light cache generated in 0.9 s (26.0 MB/s)
GPU1: Allocating DAG (2x 1.44) GB; good for epoch up to #56
GPU1: Generating DAG for epoch #54
GPU1: DAG  50%
GPU1: DAG 100%
GPU1: DAG generated in 3.2 s (452.0 MB/s)
GPU1: Unable to prepare kernels: clCreateKernel (-46)

Card is a RX 570 4 GB model. Should this not be enough memory?

    Thank you for reporting this. The -clkernel 3 option doesn't work with dual mining yet. We will try to include the turbo dual mining kernels in the final 4.1 release.

any plans for kernal 3 for fury x or vega 64?

also the setting of etth (claymore lower limti of 50 being lowered to 25?) i think can help squeeze out some more hashs..

thanks

   Unfortunately the techniques that make -clkernel 3 faster doesn't work on Fury and Vega GPUs. Can you please clarify what do you mean by etth?