Author

Topic: PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux) - page 417. (Read 784958 times)

newbie
Activity: 6
Merit: 0
Hi,

I recently installed the PhoenixMiner 2.9e, moving from Claymore where I was mining ETH to now mine PIRL and am having some issues, hoping you all can help me resolve.  I get a CUDA error 377 out of memory error on the nVidia cards (1060 3 GB),  the AMD cards work fine.

Rig:

ASRosk HTC110 Mobo
3 AMD 570 gpus
8 nVidia 1060 3 GB GPUs
8 GB RAM
250 GB HD

nvidia driver 388.59 (I've tried an older 37X.xx version, did not help)
Virtual Memory - minimum set to 60,000, max set to 160,000.
Set to mine PIRL so that the DAG is not a problem

REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM PhoenixMiner.exe -pool pirl.minerpool.net:8002 -wal YourPirlWalletAddress -pass x -worker WorkerName
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

REM IMPORTANT: Replace the ETH address with your own ETH wallet address in the -wal option (Rig001 is the name of the rig)
PhoenixMiner.exe -pool ssl://eu1.ethermine.org:5555 -pool2 ssl://us1.ethermine.org:5555 -wal 0xac1CbCB36704BD0BE0A773cDAFE98c72a70707e2.CocoSpency -eres 0 -lidag 3
pause

This rig has been successfully with claymore on ETH for several months. Then MS pushed the update that 'killed' the 1060's 3GB from being able to mine ETH.  Switched to your miner and PIRL to keep using them.

Would welcome any thoughts, I can't seem to find any solutions online.

Seems your are using your ETH wallet to mine pirl, you must have a pirl wallet.
The pool is an ETH pool, you must log in a PIRL pool

PhoenixMiner.exe -pool pirl.minerpool.net:8002  -wal USE_YOUR_PIRL_WALLET_HERE.CocoSpency -eres 0 -lidag 3

IAMTUTU - Thanks for finding my additional error!  In my haste of copying pasting between bat files, I jacked that all up.  Got that part fixed, but am getting this error.

2018.05.02:13:15:40.936: GPU2 GPU2: Generating DAG for epoch #43
2018.05.02:13:15:40.971: GPU9 CUDA error in CudaProgram.cu:102 : an illegal memory access was encountered (700)
2018.05.02:13:15:40.971: GPU9 GPU9 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:40.971: GPU10 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:40.971: GPU10 GPU10 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU11 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU11 GPU11 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU4 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU4 GPU4 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU5 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU5 GPU5 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU2 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU2 GPU2 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU7 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU7 GPU7 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU3 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU3 GPU3 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.129: GPU1 GPU1: Using new optimized OpenCL kernels (device name 'Ellesmere')
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating DAG for epoch #43 (1.34) GB
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating light cache buffer (21.4) MB; good for epoch up to #43
2018.05.02:13:15:41.184: GPU1 GPU1: Generating DAG for epoch #43


Thoughts?

See thread..
https://bitcointalksearch.org/topic/m.31966965


Appreciate the thought - I've reviewed that thread.  The problem here is that the GPUs are not OC'ed at all and the errors are thrown right after starting the miner, so the nvidia GPUs never have a chance to start mining.

Likely just the overclock setting needs to be lowered...
full member
Activity: 1120
Merit: 131
Indeed, memory errors result in general from too much OC.
jr. member
Activity: 117
Merit: 3
Hi,

I recently installed the PhoenixMiner 2.9e, moving from Claymore where I was mining ETH to now mine PIRL and am having some issues, hoping you all can help me resolve.  I get a CUDA error 377 out of memory error on the nVidia cards (1060 3 GB),  the AMD cards work fine.

Rig:

ASRosk HTC110 Mobo
3 AMD 570 gpus
8 nVidia 1060 3 GB GPUs
8 GB RAM
250 GB HD

nvidia driver 388.59 (I've tried an older 37X.xx version, did not help)
Virtual Memory - minimum set to 60,000, max set to 160,000.
Set to mine PIRL so that the DAG is not a problem

REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM PhoenixMiner.exe -pool pirl.minerpool.net:8002 -wal YourPirlWalletAddress -pass x -worker WorkerName
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

REM IMPORTANT: Replace the ETH address with your own ETH wallet address in the -wal option (Rig001 is the name of the rig)
PhoenixMiner.exe -pool ssl://eu1.ethermine.org:5555 -pool2 ssl://us1.ethermine.org:5555 -wal 0xac1CbCB36704BD0BE0A773cDAFE98c72a70707e2.CocoSpency -eres 0 -lidag 3
pause

This rig has been successfully with claymore on ETH for several months. Then MS pushed the update that 'killed' the 1060's 3GB from being able to mine ETH.  Switched to your miner and PIRL to keep using them.

Would welcome any thoughts, I can't seem to find any solutions online.

Seems your are using your ETH wallet to mine pirl, you must have a pirl wallet.
The pool is an ETH pool, you must log in a PIRL pool

PhoenixMiner.exe -pool pirl.minerpool.net:8002  -wal USE_YOUR_PIRL_WALLET_HERE.CocoSpency -eres 0 -lidag 3

IAMTUTU - Thanks for finding my additional error!  In my haste of copying pasting between bat files, I jacked that all up.  Got that part fixed, but am getting this error.

2018.05.02:13:15:40.936: GPU2 GPU2: Generating DAG for epoch #43
2018.05.02:13:15:40.971: GPU9 CUDA error in CudaProgram.cu:102 : an illegal memory access was encountered (700)
2018.05.02:13:15:40.971: GPU9 GPU9 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:40.971: GPU10 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:40.971: GPU10 GPU10 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU11 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU11 GPU11 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU4 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU4 GPU4 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU5 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU5 GPU5 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU2 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU2 GPU2 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU7 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU7 GPU7 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU3 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU3 GPU3 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.129: GPU1 GPU1: Using new optimized OpenCL kernels (device name 'Ellesmere')
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating DAG for epoch #43 (1.34) GB
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating light cache buffer (21.4) MB; good for epoch up to #43
2018.05.02:13:15:41.184: GPU1 GPU1: Generating DAG for epoch #43


Thoughts?

See thread..
https://bitcointalksearch.org/topic/m.31966965

Likely just the overclock setting needs to be lowered...
newbie
Activity: 1
Merit: 0

VirusTotal shows 14/16 AVs detecting malware in PhoenixMiner 2.9e.

Something to be concerned about? 2.9d was fine.

https://www.virustotal.com/en/file/a531b7b0bb925173d3ea2976b72f3d280f64751bdb094d5bb980553dfa85fb07/analysis/

newbie
Activity: 6
Merit: 0
Hi,

I recently installed the PhoenixMiner 2.9e, moving from Claymore where I was mining ETH to now mine PIRL and am having some issues, hoping you all can help me resolve.  I get a CUDA error 377 out of memory error on the nVidia cards (1060 3 GB),  the AMD cards work fine.

Rig:

ASRosk HTC110 Mobo
3 AMD 570 gpus
8 nVidia 1060 3 GB GPUs
8 GB RAM
250 GB HD

nvidia driver 388.59 (I've tried an older 37X.xx version, did not help)
Virtual Memory - minimum set to 60,000, max set to 160,000.
Set to mine PIRL so that the DAG is not a problem

REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM PhoenixMiner.exe -pool pirl.minerpool.net:8002 -wal YourPirlWalletAddress -pass x -worker WorkerName
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

REM IMPORTANT: Replace the ETH address with your own ETH wallet address in the -wal option (Rig001 is the name of the rig)
PhoenixMiner.exe -pool ssl://eu1.ethermine.org:5555 -pool2 ssl://us1.ethermine.org:5555 -wal 0xac1CbCB36704BD0BE0A773cDAFE98c72a70707e2.CocoSpency -eres 0 -lidag 3
pause

This rig has been successfully with claymore on ETH for several months. Then MS pushed the update that 'killed' the 1060's 3GB from being able to mine ETH.  Switched to your miner and PIRL to keep using them.

Would welcome any thoughts, I can't seem to find any solutions online.

Seems your are using your ETH wallet to mine pirl, you must have a pirl wallet.
The pool is an ETH pool, you must log in a PIRL pool

PhoenixMiner.exe -pool pirl.minerpool.net:8002  -wal USE_YOUR_PIRL_WALLET_HERE.CocoSpency -eres 0 -lidag 3

IAMTUTU - Thanks for finding my additional error!  In my haste of copying pasting between bat files, I jacked that all up.  Got that part fixed, but am getting this error.

2018.05.02:13:15:40.936: GPU2 GPU2: Generating DAG for epoch #43
2018.05.02:13:15:40.971: GPU9 CUDA error in CudaProgram.cu:102 : an illegal memory access was encountered (700)
2018.05.02:13:15:40.971: GPU9 GPU9 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:40.971: GPU10 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:40.971: GPU10 GPU10 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU11 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU11 GPU11 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU4 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU4 GPU4 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU5 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU5 GPU5 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU2 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU2 GPU2 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU7 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU7 GPU7 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.071: GPU3 CUDART error in CudaProgram.cu:127 : an illegal memory access was encountered (77)
2018.05.02:13:15:41.071: GPU3 GPU3 initMiner error: an illegal memory access was encountered
2018.05.02:13:15:41.129: GPU1 GPU1: Using new optimized OpenCL kernels (device name 'Ellesmere')
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating DAG for epoch #43 (1.34) GB
2018.05.02:13:15:41.129: GPU1 GPU1: Allocating light cache buffer (21.4) MB; good for epoch up to #43
2018.05.02:13:15:41.184: GPU1 GPU1: Generating DAG for epoch #43


Thoughts?
jr. member
Activity: 47
Merit: 1
Who can say - why sometimes speed 1 of cards drops from 31M to 25M untill restarting miner?

Overclock reset. It happens sometimes don't know why though.

I have these hash drops also but it comes back in couple of minutes. Sometime OverdriveNtool loses OC setting for one card usually, just down OC


I use PhoenixMiner for OC (AMD) and MSI Afterburner for Nvidia. Have the drops and they always recover to max hash in couple of minutes. Never required to restart or anything. One of the "why" PhoenixMiner.
full member
Activity: 1120
Merit: 131
Hi,

I recently installed the PhoenixMiner 2.9e, moving from Claymore where I was mining ETH to now mine PIRL and am having some issues, hoping you all can help me resolve.  I get a CUDA error 377 out of memory error on the nVidia cards (1060 3 GB),  the AMD cards work fine.

Rig:

ASRosk HTC110 Mobo
3 AMD 570 gpus
8 nVidia 1060 3 GB GPUs
8 GB RAM
250 GB HD

nvidia driver 388.59 (I've tried an older 37X.xx version, did not help)
Virtual Memory - minimum set to 60,000, max set to 160,000.
Set to mine PIRL so that the DAG is not a problem

REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM PhoenixMiner.exe -pool pirl.minerpool.net:8002 -wal YourPirlWalletAddress -pass x -worker WorkerName
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

REM IMPORTANT: Replace the ETH address with your own ETH wallet address in the -wal option (Rig001 is the name of the rig)
PhoenixMiner.exe -pool ssl://eu1.ethermine.org:5555 -pool2 ssl://us1.ethermine.org:5555 -wal 0xac1CbCB36704BD0BE0A773cDAFE98c72a70707e2.CocoSpency -eres 0 -lidag 3
pause

This rig has been successfully with claymore on ETH for several months. Then MS pushed the update that 'killed' the 1060's 3GB from being able to mine ETH.  Switched to your miner and PIRL to keep using them.

Would welcome any thoughts, I can't seem to find any solutions online.

Seems your are using your ETH wallet to mine pirl, you must have a pirl wallet.
The pool is an ETH pool, you must log in a PIRL pool

PhoenixMiner.exe -pool pirl.minerpool.net:8002  -wal USE_YOUR_PIRL_WALLET_HERE.CocoSpency -eres 0 -lidag 3
jr. member
Activity: 47
Merit: 1
One of my rigs stopped mining earlier today. Turned out it was Windows Defender that moved the PhoenixMiner.exe into the quarantine categorised as "Trojan:Win32/Tilken.B!cl"
Did restore the .exe and disabled Windows Defender. The threat in my opinion is Windows Defender, not the miner.

Windows Defender Details:
Trojan:Win32/Tilken.B!cl
Alert level: Severe
Status: Active
Date: 02/05/2018
Recommended action: Remove threat now.

Category: Trojan
Details: This program is dangerous and executes commands from an attacker.

Affected items:
file: C:\dev\Ethereum\PhoenixMiner_2.9e\PhoenixMiner.exe
process: pid:10100,ProcessStart:131683638871312779
You should not use any AV. included WD.
That's really bad advice.
Just configure your AV to exclude certain directories that your miners are located.


To be precise: "Disabled Windows Defender" by adding an exclusion folder for PhoenixMiner. in the Windows Defender Security center: e.g. c:\dev\Ethereum\PhoenixMiner_2.9e
newbie
Activity: 6
Merit: 0
Hi,

I recently installed the PhoenixMiner 2.9e, moving from Claymore where I was mining ETH to now mine PIRL and am having some issues, hoping you all can help me resolve.  I get a CUDA error 377 out of memory error on the nVidia cards (1060 3 GB),  the AMD cards work fine.

Rig:

ASRosk HTC110 Mobo
3 AMD 570 gpus
8 nVidia 1060 3 GB GPUs
8 GB RAM
250 GB HD

nvidia driver 388.59 (I've tried an older 37X.xx version, did not help)
Virtual Memory - minimum set to 60,000, max set to 160,000.
Set to mine PIRL so that the DAG is not a problem

REM
REM Example bat file for starting PhoenixMiner.exe to mine ETH
REM PhoenixMiner.exe -pool pirl.minerpool.net:8002 -wal YourPirlWalletAddress -pass x -worker WorkerName
REM

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

REM IMPORTANT: Replace the ETH address with your own ETH wallet address in the -wal option (Rig001 is the name of the rig)
PhoenixMiner.exe -pool ssl://eu1.ethermine.org:5555 -pool2 ssl://us1.ethermine.org:5555 -wal 0xac1CbCB36704BD0BE0A773cDAFE98c72a70707e2.CocoSpency -eres 0 -lidag 3
pause

This rig has been successfully with claymore on ETH for several months. Then MS pushed the update that 'killed' the 1060's 3GB from being able to mine ETH.  Switched to your miner and PIRL to keep using them.

Would welcome any thoughts, I can't seem to find any solutions online.
member
Activity: 277
Merit: 23
Who can say - why sometimes speed 1 of cards drops from 31M to 25M untill restarting miner?

Overclock reset. It happens sometimes don't know why though.

I have these hash drops also but it comes back in couple of minutes. Sometime OverdriveNtool loses OC setting for one card usually, just down OC
newbie
Activity: 23
Merit: 0
Who can say - why sometimes speed 1 of cards drops from 31M to 25M untill restarting miner?

Overclock reset. It happens sometimes don't know why though.
sr. member
Activity: 1484
Merit: 253
Who can say - why sometimes speed 1 of cards drops from 31M to 25M untill restarting miner?
jr. member
Activity: 170
Merit: 6
One of my rigs stopped mining earlier today. Turned out it was Windows Defender that moved the PhoenixMiner.exe into the quarantine categorised as "Trojan:Win32/Tilken.B!cl"
Did restore the .exe and disabled Windows Defender. The threat in my opinion is Windows Defender, not the miner.

Windows Defender Details:
Trojan:Win32/Tilken.B!cl
Alert level: Severe
Status: Active
Date: 02/05/2018
Recommended action: Remove threat now.

Category: Trojan
Details: This program is dangerous and executes commands from an attacker.

Affected items:
file: C:\dev\Ethereum\PhoenixMiner_2.9e\PhoenixMiner.exe
process: pid:10100,ProcessStart:131683638871312779
You should not use any AV. included WD.
That's really bad advice.
Just configure your AV to exclude certain directories that your miners are located.
newbie
Activity: 51
Merit: 0
One of my rigs stopped mining earlier today. Turned out it was Windows Defender that moved the PhoenixMiner.exe into the quarantine categorised as "Trojan:Win32/Tilken.B!cl"
Did restore the .exe and disabled Windows Defender. The threat in my opinion is Windows Defender, not the miner.

Windows Defender Details:
Trojan:Win32/Tilken.B!cl
Alert level: Severe
Status: Active
Date: 02/05/2018
Recommended action: Remove threat now.

Category: Trojan
Details: This program is dangerous and executes commands from an attacker.

Affected items:
file: C:\dev\Ethereum\PhoenixMiner_2.9e\PhoenixMiner.exe
process: pid:10100,ProcessStart:131683638871312779
You should not use any AV. included WD.
newbie
Activity: 2
Merit: 0
jr. member
Activity: 47
Merit: 1
Core clock should be use to lower the load on memmory but should never get over +150, and this also lower the power to get the card running between 80 and 100 watts with reduced power limit between 50 and 60%

Here you have a good guide how to overclock  nvidia cards https://mining.help/nvidia-ethereum-mining/

In my opinion the guide above (and other guides on the web) are 80% great but inaccurate at 20%. A great start for newcomers - it helped me to get the whole picture when I started to build first rig, but inaccurate at many aspects and details.
Advice: Proceed with caution.

Again: My personal opinion.
jr. member
Activity: 47
Merit: 1
One of my rigs stopped mining earlier today. Turned out it was Windows Defender that moved the PhoenixMiner.exe into the quarantine categorised as "Trojan:Win32/Tilken.B!cl"
Did restore the .exe and disabled Windows Defender. The threat in my opinion is Windows Defender, not the miner.

Windows Defender Details:
Trojan:Win32/Tilken.B!cl
Alert level: Severe
Status: Active
Date: 02/05/2018
Recommended action: Remove threat now.

Category: Trojan
Details: This program is dangerous and executes commands from an attacker.

Affected items:
file: C:\dev\Ethereum\PhoenixMiner_2.9e\PhoenixMiner.exe
process: pid:10100,ProcessStart:131683638871312779
jr. member
Activity: 47
Merit: 1
hashrate with gtx 1070 ?

These are my stable settings:
Driver: 391.01
Using MSI Afterburner set as:
  Core Voltage (%): 0
  Power Limit (%): 55
  Temp Limit (%): 83 (linked with Power Limit)
  Core Clock (MHz): -200
  Memory Clock (MHz): 635


Is your core clock set to negative 200 mhz?  I'm getting similar hash rates with my MSI GTX 1070s, (just two for the time being.)  Here are my afterburner settings:

Core Voltage: 0%
Power: 65%
Temp: 67 degrees Centigrade (linked with power)
Core Clock: +99
Mem Clock: +650


Yes, I run my core clock as low as possible to preserve power, lower the overall GPU temperature. Ethash PoW is memory hard. High core clock has very little impact on hash rate. So no real gain increasing the core clock.

My new settings (stable for the last 24h) are:

  Core Voltage (%): +0
  Power Limit (%): 58
  Temp Limit (%): 83 (linked with Power Limit)
  Core Clock (MHz): -200 (minus two hundred)
  Memory Clock (MHz): +685

With PhoenixMiner stats are:
Hash:
1: 32.114 MH/s (4419) 2: 32.113 MH/s (4435) 3: 32.098 MH/s (4351) 4: 32.116 MH/s (4478/3)
6: 32.128 MH/s (4469) 7: 32.119 MH/s (4390) 12: 32.120 MH/s (4489)

Temperature:
GPU1: 58C 38%, GPU2: 60C 41%, GPU3: 57C 36%, GPU4: 56C 35%, GPU6: 58C 39%, GPU7: 60C 41%, GPU12: 57C 36%, GPU13: 53C 0%

While running on the default configuration I would be getting around 26 MH/s while the GPU temperature would stabilise at 72oC.
By reducing the core/power limit both the GPU temperature and power usage is reduced.

In the above stats I have removed the RX580 figures as this reply is focusing on nvidia GTX 1070.
The above GPUs are all Gigabyte - GV-N1070WF2OC-8GD (GIGABYTE GTX 1070 8GB GDDR5)
newbie
Activity: 129
Merit: 0
I think the "Every 5 seconds" shown speed or the average 5 min. are wrong.
With my 2 x RX580 & -clKernel 0
I have been looking carefully for some 15 min. periods, at the screen I see:

70 % of time 64+ MH/s (rarely up to 65)
15 % 62 MH/s
15 % 58.1-58.2

The average 5 min. shown is 57.9 to 58 after 15 min., lowest than the lowest 5 seconds speed.
Also the average goes lower within time while the 5 sec. are the same.
I think the same happens with Kernel 1 but differences are small & it is harder to know for sure, I have not tested it carefully yet.

Looks like the 5 min. average is metering the lowest H/s or I don't know what.

More:

I checked it again this time for more than 30 min.

The lowest 5 sec. have never been below 58.15 (Most of time between 58.2+ & 64.7+) while the highest average 5 min. have never reached 58.1 MH/s (58.08 at most, usually below 58).
So, something is not OK.

newbie
Activity: 15
Merit: 0
Miner must be agressive if you want to get the the benefit from mining..
Jump to: