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

[Owl182]

Hi,

Ok, i need help, everything runs fine until it says there is not enough Vram and DAG auto allocation stops, then it allocates incrementally to 100% but then seems to crash, red text all over the place... what can i do ?

(dont say buy another card...)

[1713871186]

1713871186

[1713871186]

1713871186

[1713871186]

1713871186

[Gaelyon]

Hello Owl182, to mine Ethereum you need a graphic card with more than 4gb of vram (usually 6+gb cards are fine). The error log seem to indicate that's your problem. And yes, in that case there's no other solution than buying a new card with 6+vram. Sorry.

[Owl182]

Hi,

Yeah ive learned as much, however, it is running on Kryptex on 5.5c just fine at 25.192MH/s ? dunno what the parameters are though so cant run Phoenix on its own

[djshadow21922]

Im currently using a 6900xt card which is new to me.  Before this card I was using BetterHash software.  Which was great.   I love betterhash.

But I cant seem to use the 6900xt with Phenox miner to mine ETh or ETC.  Betterhash has worked with me and we have changed a few settings within Betterhash and also with my Virtual Memory.  But nothing has worked. And they suggest to reach out to this form.

For the time being I'm using NiceHash NBMiner and it seems to be working fine.   But I want Betterhash


Phoenix Miner 5.5c Windows/msvc - Release build
-----------------------------------------------

Unknown OpenCL driver version! Hashrate and stale shares may suffer
OpenCL platform: OpenCL 2.1 AMD-APP (3240.6)
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
Eth: the pool list contains 2 pools (2 from command-line)
Eth: primary pool: 88.198.11.177:55456
Starting GPU mining
GPU0: AMD driver 21.4.1
Eth: Connecting to ethash pool 88.198.11.177:55456 (proto: EthProxy)
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Eth: Connected to ethash pool 88.198.11.177:55456 (88.198.11.177)
Listening for CDM remote manager at port 33455 in read-only mode
Eth: New job #a6eafa3b from 88.198.11.177:55456; diff: 4000MH
GPU0: Starting up... (0)
GPU0: Generating etchash light cache for epoch #212

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

Light cache generated in 3.6 s (11.9 MB/s)
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

GPU0: Free VRAM: 15.922 GB; used: 0.063 GB
GPU0: Allocating DAG for epoch #212 (2.66) GB
GPU0: Generating DAG for epoch #212
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00
GPU0: DAG generated in 1.5 s (1812.1 MB/s)
GPU0: Using Ethash OCL kernels (gfx1030; -clkernel 1 -clnew 0)
GPU0: no -gt value specified, switching to auto-tune
GPU0: starting auto-tune process
GPU0: clSetKernelArg (-48)
Fatal error detected. Restarting.

*** 0:00 *** 5/17 20:50 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

[Targonis]

You are running Phoenix 5.5c while the latest version is 5.6d(real version, check the first post in this thread for confirmation).   Anything newer than Radeon Software 21.3.2 is NOT compatible with Phoenix 5.5c.   

It's as simple as that, AMD made some significant changes to the drivers starting with 21.4.1.

Im currently using a 6900xt card which is new to me.  Before this card I was using BetterHash software.  Which was great.   I love betterhash.

But I cant seem to use the 6900xt with Phenox miner to mine ETh or ETC.  Betterhash has worked with me and we have changed a few settings within Betterhash and also with my Virtual Memory.  But nothing has worked. And they suggest to reach out to this form.

For the time being I'm using NiceHash NBMiner and it seems to be working fine.   But I want Betterhash


Phoenix Miner 5.5c Windows/msvc - Release build
-----------------------------------------------

Unknown OpenCL driver version! Hashrate and stale shares may suffer
OpenCL platform: OpenCL 2.1 AMD-APP (3240.6)
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
Eth: the pool list contains 2 pools (2 from command-line)
Eth: primary pool: 88.198.11.177:55456
Starting GPU mining
GPU0: AMD driver 21.4.1
Eth: Connecting to ethash pool 88.198.11.177:55456 (proto: EthProxy)
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Eth: Connected to ethash pool 88.198.11.177:55456 (88.198.11.177)
Listening for CDM remote manager at port 33455 in read-only mode
Eth: New job #a6eafa3b from 88.198.11.177:55456; diff: 4000MH
GPU0: Starting up... (0)
GPU0: Generating etchash light cache for epoch #212

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

Light cache generated in 3.6 s (11.9 MB/s)
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

GPU0: Free VRAM: 15.922 GB; used: 0.063 GB
GPU0: Allocating DAG for epoch #212 (2.66) GB
GPU0: Generating DAG for epoch #212
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00
GPU0: DAG generated in 1.5 s (1812.1 MB/s)
GPU0: Using Ethash OCL kernels (gfx1030; -clkernel 1 -clnew 0)
GPU0: no -gt value specified, switching to auto-tune
GPU0: starting auto-tune process
GPU0: clSetKernelArg (-48)
Fatal error detected. Restarting.

*** 0:00 *** 5/17 20:50 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

[snappleapple100]

PM 5.6d just randomly closed on me, was happening with PM 5.5c is this more related to my hardware making it close or its just on PM ? any one have this issue ?

[Owl182]

Hi,

Has anyone got a "working" breakdown of this file ? i have the file itself but cant seem to get the parameters right, i'm also new to this so its pretty difficult...

mine_eth_4G.bat (one that works with Phoenix 5.6d that is...)

Thanks in advance...

[djshadow21922]

Thank you so much Targonis.  I have referred this information back over to BetterHash so they can get their Phoenix Miner plug in updated.  They made it sound like it was Phoenix issue and not their issue. 

You are running Phoenix 5.5c while the latest version is 5.6d(real version, check the first post in this thread for confirmation).   Anything newer than Radeon Software 21.3.2 is NOT compatible with Phoenix 5.5c.   

It's as simple as that, AMD made some significant changes to the drivers starting with 21.4.1.

Im currently using a 6900xt card which is new to me.  Before this card I was using BetterHash software.  Which was great.   I love betterhash.

But I cant seem to use the 6900xt with Phenox miner to mine ETh or ETC.  Betterhash has worked with me and we have changed a few settings within Betterhash and also with my Virtual Memory.  But nothing has worked. And they suggest to reach out to this form.

For the time being I'm using NiceHash NBMiner and it seems to be working fine.   But I want Betterhash


Phoenix Miner 5.5c Windows/msvc - Release build
-----------------------------------------------

Unknown OpenCL driver version! Hashrate and stale shares may suffer
OpenCL platform: OpenCL 2.1 AMD-APP (3240.6)
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
Eth: the pool list contains 2 pools (2 from command-line)
Eth: primary pool: 88.198.11.177:55456
Starting GPU mining
GPU0: AMD driver 21.4.1
Eth: Connecting to ethash pool 88.198.11.177:55456 (proto: EthProxy)
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Eth: Connected to ethash pool 88.198.11.177:55456 (88.198.11.177)
Listening for CDM remote manager at port 33455 in read-only mode
Eth: New job #a6eafa3b from 88.198.11.177:55456; diff: 4000MH
GPU0: Starting up... (0)
GPU0: Generating etchash light cache for epoch #212

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

Light cache generated in 3.6 s (11.9 MB/s)
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00

*** 0:00 *** 5/17 20:49 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

GPU0: Free VRAM: 15.922 GB; used: 0.063 GB
GPU0: Allocating DAG for epoch #212 (2.66) GB
GPU0: Generating DAG for epoch #212
Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00
GPU0: DAG generated in 1.5 s (1812.1 MB/s)
GPU0: Using Ethash OCL kernels (gfx1030; -clkernel 1 -clnew 0)
GPU0: no -gt value specified, switching to auto-tune
GPU0: starting auto-tune process
GPU0: clSetKernelArg (-48)
Fatal error detected. Restarting.

*** 0:00 *** 5/17 20:50 **************************************
Eth: Mining ETC on 88.198.11.177:55456 for 0:00
Available GPUs for mining:
GPU0: AMD Radeon RX 6900 XT (pcie 3), OpenCL 2.0, 16 GB VRAM, 40 CUs
GPU0: 48C 0% 12W
GPUs power: 12.0 W
Current -gt 15
Eth: Accepted shares 0 (0 stales), rejected shares 0 (0 stales)
Eth: Incorrect shares 0 (0.00%), est. stales percentage 0.00%
Eth: Average speed (5 min): 0.000 MH/s

[Seph213]

im noticing on phoenix miner 5.6d alot more stale shares then using previous 5.5c any body else see a increase im using nvidia cards

This happened to me after activating a new feature in windows. it was on SYSTEM: Monitor (first item on top)- graphicsettings (the last one)- hardwarebased GPU (?).
Dont turn this on. Turn OFF.

I have it turned off and still get more stale shares than before .Anyone has an idea why this suddenly happens ? Been having 1% stale rate for months now its to 3-4 with no change in hardware or miner

[doctorman]

@PhoenixMiner

1)   How can I find out with GPU number is connected to which PCIE port?

I have 14 GPU on a rig  using  ASUS B250 Mininging expert , Each pcie is numbered on the motherboard
after a while  and  GPU crashes restarts ,  I noticed my GPU numbers do not match to the PCIE ports I set them originally

It is hell to find out which GPU is which...  any help in identifying  GPU numbers is greatly appreciated


2)  If I want same settings for all RTX 3070s and  special settings for  ie.  RTX3090 on the same rig how do I have to set that in the string of phoenix miner  considering the GPU numbers can switch or are hard to find?

running in windows

[Binary100100]

I give up on trying to fix this issue on my own. I'm turning to the community for help.
Here's the deal.
HiveOS running Phoenixminer 5.6d with 6 RX 5700XT Sapphire Pulse cards.
They're running really hot! They all have the exact same memory type (Micron) and are all flashed with the exact same bios.
My tuning commands are simply: -tt 90 -fanmin 95
My clocks are as follows: (these seem to be the most stable and optimal during the Spring/Summer months)
FAN 99
CORE 1400
VDD 800
MEM 910
MVDD 1300
VDDCI 800

Now some cards are running at 57MH/s. That's fine. That's what I want... at minimum. Those temperatures are 68 Core and 102 Memory.
But others are 50MH/s and temperatures are 74 Core and 106 Memory. So I'm chalking it up to thermal throttling.

Okay. So let's boost the fan speeds... but the fans are already set to 99%. But HiveOS is telling me that the fans are only running at speeds between 81% - 88%. No where near the 99% that I've set the command (-tt 90 -fanmin 95) to.
Where did I screw up?

Drivers 20.40
Kernel 5.4.80
OS Ver 0.6-203@210519

[dafuq8008]

I give up on trying to fix this issue on my own. I'm turning to the community for help.
Here's the deal.
HiveOS running Phoenixminer 5.6d with 6 RX 5700XT Sapphire Pulse cards.
They're running really hot! They all have the exact same memory type (Micron) and are all flashed with the exact same bios.
My tuning commands are simply: -tt 90 -fanmin 95
My clocks are as follows: (these seem to be the most stable and optimal during the Spring/Summer months)
FAN 99
CORE 1400
VDD 800
MEM 910
MVDD 1300
VDDCI 800

Now some cards are running at 57MH/s. That's fine. That's what I want... at minimum. Those temperatures are 68 Core and 102 Memory.
But others are 50MH/s and temperatures are 74 Core and 106 Memory. So I'm chalking it up to thermal throttling.

Okay. So let's boost the fan speeds... but the fans are already set to 99%. But HiveOS is telling me that the fans are only running at speeds between 81% - 88%. No where near the 99% that I've set the command (-tt 90 -fanmin 95) to.
Where did I screw up?

Drivers 20.40
Kernel 5.4.80
OS Ver 0.6-203@210519

You're really pushing the limits of those 5700 xts.  I have 6 5700 xt red devils running:

1350 gpu/850 volts/885 memory, and the memory timings from 1500 MHz.  I'm getting about 54.7 mh/s.  Fans run ~24% with temps in the low to mid 50s for the gpu (I'm on ubuntu so I'm not sure of the memory temp).  They are rock solid.  That is slower than your top, but about the same considering your throttling.  

It's up to you, but your only option is to scale back your memory or open them up and put in better thermal pads.  (I'd suspect your cards would last longer too.)

It's all about winning or losing the silicon lottery. It's hard to treat them all the same.

[Binary100100]

I give up on trying to fix this issue on my own. I'm turning to the community for help.
Here's the deal.
HiveOS running Phoenixminer 5.6d with 6 RX 5700XT Sapphire Pulse cards.
They're running really hot! They all have the exact same memory type (Micron) and are all flashed with the exact same bios.
My tuning commands are simply: -tt 90 -fanmin 95
My clocks are as follows: (these seem to be the most stable and optimal during the Spring/Summer months)
FAN 99
CORE 1400
VDD 800
MEM 910
MVDD 1300
VDDCI 800

Now some cards are running at 57MH/s. That's fine. That's what I want... at minimum. Those temperatures are 68 Core and 102 Memory.
But others are 50MH/s and temperatures are 74 Core and 106 Memory. So I'm chalking it up to thermal throttling.

Okay. So let's boost the fan speeds... but the fans are already set to 99%. But HiveOS is telling me that the fans are only running at speeds between 81% - 88%. No where near the 99% that I've set the command (-tt 90 -fanmin 95) to.
Where did I screw up?

Drivers 20.40
Kernel 5.4.80
OS Ver 0.6-203@210519

You're really pushing the limits of those 5700 xts.  I have 6 5700 xt red devils running:

1350 gpu/850 volts/885 memory, and the memory timings from 1500 MHz.  I'm getting about 54.7 mh/s.  Fans run ~24% with temps in the low to mid 50s for the gpu (I'm on ubuntu so I'm not sure of the memory temp).  They are rock solid.  That is slower than your top, but about the same considering your throttling.  

It's up to you, but your only option is to scale back your memory or open them up and put in better thermal pads.  (I'd suspect your cards would last longer too.)

It's all about winning or losing the silicon lottery. It's hard to treat them all the same.

Thanks for such a prompt response.
They were rock solid during the winter months as they were in my basement but obviously when the ambient temperature started climbing so did the amount of crashes.
The issue here really isn't the stability but perhaps the configuration. Bottom line is that I'd really like to figure out why my fans won't (or can't?) reach in the 90% - 100% range despite my commands in Phoenix to do so.
I'm telling PM "Make the fans go max speed!"
Phoenix says "Fans...... go almost max speed. Four-fifths outta do it."
Then PM later says, "Oh look... your thermal throttling."

I suppose that's my issue.

[dafuq8008]

I give up on trying to fix this issue on my own. I'm turning to the community for help.
Here's the deal.
HiveOS running Phoenixminer 5.6d with 6 RX 5700XT Sapphire Pulse cards.
They're running really hot! They all have the exact same memory type (Micron) and are all flashed with the exact same bios.
My tuning commands are simply: -tt 90 -fanmin 95
My clocks are as follows: (these seem to be the most stable and optimal during the Spring/Summer months)
FAN 99
CORE 1400
VDD 800
MEM 910
MVDD 1300
VDDCI 800

Now some cards are running at 57MH/s. That's fine. That's what I want... at minimum. Those temperatures are 68 Core and 102 Memory.
But others are 50MH/s and temperatures are 74 Core and 106 Memory. So I'm chalking it up to thermal throttling.

Okay. So let's boost the fan speeds... but the fans are already set to 99%. But HiveOS is telling me that the fans are only running at speeds between 81% - 88%. No where near the 99% that I've set the command (-tt 90 -fanmin 95) to.
Where did I screw up?

Drivers 20.40
Kernel 5.4.80
OS Ver 0.6-203@210519

You're really pushing the limits of those 5700 xts.  I have 6 5700 xt red devils running:

1350 gpu/850 volts/885 memory, and the memory timings from 1500 MHz.  I'm getting about 54.7 mh/s.  Fans run ~24% with temps in the low to mid 50s for the gpu (I'm on ubuntu so I'm not sure of the memory temp).  They are rock solid.  That is slower than your top, but about the same considering your throttling. 

It's up to you, but your only option is to scale back your memory or open them up and put in better thermal pads.  (I'd suspect your cards would last longer too.)

It's all about winning or losing the silicon lottery. It's hard to treat them all the same.

Thanks for such a prompt response.
They were rock solid during the winter months as they were in my basement but obviously when the ambient temperature started climbing so did the amount of crashes.
The issue here really isn't the stability but perhaps the configuration. Bottom line is that I'd really like to figure out why my fans won't (or can't?) reach in the 90% - 100% range despite my commands in Phoenix to do so.
I'm telling PM "Make the fans go max speed!"
Phoenix says "Fans...... go almost max speed. Four-fifths outta do it."
Then PM later says, "Oh look... your thermal throttling."

I suppose that's my issue.

Ahh.. I gotcha.  Truthfully I had a lot of problems getting software overrides to work and then be stable with my Red Devils.  That's through windows and Ubuntu, but I had more success in windows.  That's what led me down the Bios route.

You should be able to lock the fan speed in the bios.  You could lock it there and not try to go through software to do it.  I'm sure that'll at least allow you to max your fans no matter what the vram temps are.

https://www.igorslab.de/en/red-bios-editor-and-morepowertool-adjust-and-optimize-your-vbios-and-even-more-stable-overclocking-navi-unlimited/2/

[Binary100100]

I give up on trying to fix this issue on my own. I'm turning to the community for help.
Here's the deal.
HiveOS running Phoenixminer 5.6d with 6 RX 5700XT Sapphire Pulse cards.
They're running really hot! They all have the exact same memory type (Micron) and are all flashed with the exact same bios.
My tuning commands are simply: -tt 90 -fanmin 95
My clocks are as follows: (these seem to be the most stable and optimal during the Spring/Summer months)
FAN 99
CORE 1400
VDD 800
MEM 910
MVDD 1300
VDDCI 800

Now some cards are running at 57MH/s. That's fine. That's what I want... at minimum. Those temperatures are 68 Core and 102 Memory.
But others are 50MH/s and temperatures are 74 Core and 106 Memory. So I'm chalking it up to thermal throttling.

Okay. So let's boost the fan speeds... but the fans are already set to 99%. But HiveOS is telling me that the fans are only running at speeds between 81% - 88%. No where near the 99% that I've set the command (-tt 90 -fanmin 95) to.
Where did I screw up?

Drivers 20.40
Kernel 5.4.80
OS Ver 0.6-203@210519

You're really pushing the limits of those 5700 xts.  I have 6 5700 xt red devils running:

1350 gpu/850 volts/885 memory, and the memory timings from 1500 MHz.  I'm getting about 54.7 mh/s.  Fans run ~24% with temps in the low to mid 50s for the gpu (I'm on ubuntu so I'm not sure of the memory temp).  They are rock solid.  That is slower than your top, but about the same considering your throttling. 

It's up to you, but your only option is to scale back your memory or open them up and put in better thermal pads.  (I'd suspect your cards would last longer too.)

It's all about winning or losing the silicon lottery. It's hard to treat them all the same.

Thanks for such a prompt response.
They were rock solid during the winter months as they were in my basement but obviously when the ambient temperature started climbing so did the amount of crashes.
The issue here really isn't the stability but perhaps the configuration. Bottom line is that I'd really like to figure out why my fans won't (or can't?) reach in the 90% - 100% range despite my commands in Phoenix to do so.
I'm telling PM "Make the fans go max speed!"
Phoenix says "Fans...... go almost max speed. Four-fifths outta do it."
Then PM later says, "Oh look... your thermal throttling."

I suppose that's my issue.

Ahh.. I gotcha.  Truthfully I had a lot of problems getting software overrides to work and then be stable with my Red Devils.  That's through windows and Ubuntu, but I had more success in windows.  That's what led me down the Bios route.

You should be able to lock the fan speed in the bios.  You could lock it there and not try to go through software to do it.  I'm sure that'll at least allow you to max your fans no matter what the vram temps are.

https://www.igorslab.de/en/red-bios-editor-and-morepowertool-adjust-and-optimize-your-vbios-and-even-more-stable-overclocking-navi-unlimited/2/

Yep. Tried that already too. That's why they're all flashed with the same bios.

[dafuq8008]

Maybe you did this, but remove the Phoenix Miner fan settings in your command options and let it just read your BIOS settings.  Perhaps it's creating some kind of conflict.

If you have the BIOS how you want you, you shouldn't need any Phoenix settings.

Other than that, I'd check your driver and see if there are fan issues with the version you are using.  I'm not sure if this applies to you, but I came across this.  Basically this driver was reporting slower fan speeds than were actually running.

https://community.amd.com/t5/drivers-software/20-2-2-fan-pwm-setting-not-going-lower-than-26/m-p/352271#M110315

Sorry man.. I wish I could be more help.

[Iamtutut]

Just updated to the lastest version (5.6d), with a 5600XT.

Mining for slightly over 3 hours.

Consistent 42MH/s miner side, couldn't reach this speed with the same settings with 5.2e.
Pool side: 38-38,1MH/s (NH)
Stales: 4/1122
Errors: 0
GPU wattage according to the driver: 106W. Dunno if it's a bug, couldn't get the GPU running @800mV before (Win 10), always stuck @900mV minimum. Power draw was always 120-125W before.

Will see if this jkeeps going on.

Much less stale than 5.2e, less power draw, a little faster. I'll keep running 5.6d

[Binary100100]

Maybe you did this, but remove the Phoenix Miner fan settings in your command options and let it just read your BIOS settings.  Perhaps it's creating some kind of conflict.

If you have the BIOS how you want you, you shouldn't need any Phoenix settings.

Other than that, I'd check your driver and see if there are fan issues with the version you are using.  I'm not sure if this applies to you, but I came across this.  Basically this driver was reporting slower fan speeds than were actually running.

https://community.amd.com/t5/drivers-software/20-2-2-fan-pwm-setting-not-going-lower-than-26/m-p/352271#M110315

Sorry man.. I wish I could be more help.

It's cool. Good to know that I'm not the only one scratching their head over it.

[hdowar]

Can someone tells me why is my log shows 15-20% stales? It's only showing on Phoenixminer, but other miners doesn't says that? Am I really losing that much shares?
On ethermine pool dashboard, it only shows 1-2% hourly. >15% reported locally is very weird? This is a bug? FYI, Ping test shows 7ms and no packet lost, connected to closest server, and tried no overclock, still the same.
People been telling me to switch to other miner, but other miners just simply not reporting stale shares at all. Please advice. Or maybe developer can explain what happened.
Thank you!!!

Part of my Log:
2021.05.14:08:30:16.821: main Eth: Mining ETH on ssl://us2.ethermine.org:5555 for 12:28
2021.05.14:08:30:16.821: main Eth speed: 58.902 MH/s, shares: 594/0/5, time: 12:28
2021.05.14:08:30:16.821: main Eth: Accepted shares 594 (91 stales), rejected shares 0 (0 stales)
2021.05.14:08:30:16.821: main Eth: Incorrect shares 5 (0.83%), est. stales percentage 15.32%
2021.05.14:08:30:16.821: main Eth: Maximum difficulty of found share: 147.8 TH (!!!)

[dazmann]

Can someone tells me why is my log shows 15-20% stales? It's only showing on Phoenixminer, but other miners doesn't says that? Am I really losing that much shares?
On ethermine pool dashboard, it only shows 1-2% hourly. >15% reported locally is very weird? This is a bug? FYI, Ping test shows 7ms and no packet lost, connected to closest server, and tried no overclock, still the same.
People been telling me to switch to other miner, but other miners just simply not reporting stale shares at all. Please advice. Or maybe developer can explain what happened.
Thank you!!!

Part of my Log:
2021.05.14:08:30:16.821: main Eth: Mining ETH on ssl://us2.ethermine.org:5555 for 12:28
2021.05.14:08:30:16.821: main Eth speed: 58.902 MH/s, shares: 594/0/5, time: 12:28
2021.05.14:08:30:16.821: main Eth: Accepted shares 594 (91 stales), rejected shares 0 (0 stales)
2021.05.14:08:30:16.821: main Eth: Incorrect shares 5 (0.83%), est. stales percentage 15.32%
2021.05.14:08:30:16.821: main Eth: Maximum difficulty of found share: 147.8 TH (!!!)

phoenixminer will show the percentage of incorrect/stale shares for the time you are running the software, on ethermine it will show you for the hour/day. You gotta find how many incorrect/stale hashes are being reported on ethermine by the same period you're running the software then you check for discrepancies.