Topic: Claymore&#039;s Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 636. (Read 6590565 times)

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

This does not work for me, after changing the kernel my hashrate is dancing between 25 and 29 Sad

Quote from: fgarpe on August 29, 2017, 01:44:50 PM

you need to fix your clk level. i do this with roc-smi
look here https://github.com/RadeonOpenCompute/ROC-smi

I was using rocm-smi before installing this kernel with a custom bios to underclock-undervolt my card, I used this command in my script before launching Claymore with no changes in the hasrate dancing

Code:

sudo /opt/rocm/bin/rocm-smi --setsclk 0 1 2 3

My clock is at level 3 when Claymore is running

Code:

sudo /opt/rocm/bin/rocm-smi -c

==================== ROCm System Management Interface ====================
================================================================================
GPU[0] : GPU Clock Level: 3 (1145Mhz)
GPU[0] : GPU Memory Clock Level: 2 (2150Mhz)
================================================================================
==================== End of ROCm SMI Log ====================

How do you fix your clk level?, what command do you launch?...

Regard

I use this:
echo yes | /opt/rocm/bin/rocm-smi/rocm-smi --setsclk 4

Same result... sometimes 29, sometimes 24, sometimes 27 :´(

Hytloth

full member

Activity: 199

Merit: 100

thanks my friend. because good miner for evga 1070

cpunkt

newbie

Activity: 6

Merit: 0

Quote from: cpunkt on August 29, 2017, 10:56:32 AM

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

This does not work for me, after changing the kernel my hashrate is dancing between 25 and 29 Sad

you need to fix your clk level. i do this with roc-smi
look here https://github.com/RadeonOpenCompute/ROC-smi

I was using rocm-smi before installing this kernel with a custom bios to underclock-undervolt my card, I used this command in my script before launching Claymore with no changes in the hasrate dancing

Code:

sudo /opt/rocm/bin/rocm-smi --setsclk 0 1 2 3

My clock is at level 3 when Claymore is running

Code:

sudo /opt/rocm/bin/rocm-smi -c

==================== ROCm System Management Interface ====================
================================================================================
GPU[0] : GPU Clock Level: 3 (1145Mhz)
GPU[0] : GPU Memory Clock Level: 2 (2150Mhz)
================================================================================
==================== End of ROCm SMI Log ====================

How do you fix your clk level?, what command do you launch?...

Regard

I use this:
echo yes | /opt/rocm/bin/rocm-smi/rocm-smi --setsclk 4

nshyamkrishna

member

Activity: 72

Merit: 11

Quote from: uDwcHYO on August 29, 2017, 02:08:51 PM

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

Its Working.. Back to normal Hashrate and stable on SMOS..

nicog5

newbie

Activity: 64

Merit: 0

@Claymore
Will the ASM mode for ETH+LBC work also for Nvidia GTX1070? Thanks

NobleGhost

newbie

Activity: 25

Merit: 0

@Claymore, take a look on this nice SCAM utilizing Claymore - https://bitcointalksearch.org/topic/--2120837

Evan

hero member

Activity: 507

Merit: 500

Need some help, I have 11 cards in a system. I can run the 7 Nvidia with no issues. And I can run 3 of the 5 AMD cards with no issues, And I can run each of the 5 AMD cards solo, but the other 2 are allowed to run with other cards it crashes my GUI. And Apparently windows but the remote client stay alive during this crash because I can DC and reconnect to it but its just a black screen.

Are there and logs that can upload to trouble shoot the issue?

kpatrik

newbie

Activity: 7

Merit: 0

Is it possible that in the future the memory gddr5x will give a performance boost?
Is it possible to optimize the algorithm? Or a problem in latency and it not fixable?

Yes. I've heard about a mining operation system that uses claymore dual miner and maximizes the performance of the rigs.
Check it out: http://minerone.net

uDwcHYO

full member

Activity: 363

Merit: 100

Is it possible that in the future the memory gddr5x will give a performance boost? There must be a difference between different types of memory. by technical characteristics it is 2 times more productive.
Is it possible to optimize the algorithm? Or a problem in latency and it not fixable?

fgarpe

newbie

Activity: 8

Merit: 0

Quote from: cpunkt on August 29, 2017, 10:56:32 AM

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

This does not work for me, after changing the kernel my hashrate is dancing between 25 and 29 Sad

Quote from: NoHash on August 29, 2017, 09:49:05 AM

you need to fix your clk level. i do this with roc-smi
look here https://github.com/RadeonOpenCompute/ROC-smi

I was using rocm-smi before installing this kernel with a custom bios to underclock-undervolt my card, I used this command in my script before launching Claymore with no changes in the hasrate dancing

Code:

sudo /opt/rocm/bin/rocm-smi --setsclk 0 1 2 3

My clock is at level 3 when Claymore is running

Code:

sudo /opt/rocm/bin/rocm-smi -c

==================== ROCm System Management Interface ====================
================================================================================
GPU[0] : GPU Clock Level: 3 (1145Mhz)
GPU[0] : GPU Memory Clock Level: 2 (2150Mhz)
================================================================================
==================== End of ROCm SMI Log ====================

How do you fix your clk level?, what command do you launch?...

Regard

Odetas

newbie

Activity: 42

Merit: 0

Quote from: jackbox on August 29, 2017, 09:35:16 AM

Quote from: NoHash on August 29, 2017, 09:49:05 AM

I am on NiceHash and ONLY have Eth/Dag checked

Nicehash will change between any Ethash coin as they see fit. Not under your control. You pick the algo, not the coin. The people that pay Nicehash pick the coin they want to mine.

NiceHash can only mine those algo's that I have checked on my system,

ETH: nicehash stratum mode
"-allpools" option is set, default pools can be used for devfee, check "Readme" file for details.
Watchdog enabled
Remote management is enabled on port 5000

DCR: Stratum - connecting to 'decred.usa.nicehash.com' <198.23.94.139> port 3354
ETH: Stratum - Connected (daggerhashimoto.usa.nicehash.com:3353)
DCR: Stratum - Connected (decred.usa.nicehash.com:3354)
ETH: Authorized
Setting DAG epoch #2...
Setting DAG epoch #2 for GPU0
Create GPU buffer for GPU0
GPU0 DAG creation time - 1704 ms
Setting DAG epoch #2 for GPU0 done

yiimp is borderline nonsense with the amount of errors and downtimes. Can you do your 2nd dual coin on dual mine on Nicehash for a given algo?

scryptr

legendary

Activity: 1797

Merit: 1028

Quote from: jackbox on August 29, 2017, 09:35:16 AM

I am on NiceHash and ONLY have Eth/Dag checked

Nicehash will change between any Ethash coin as they see fit. Not under your control. You pick the algo, not the coin. The people that pay Nicehash pick the coin they want to mine.

NiceHash can only mine those algo's that I have checked on my system,

ETH: nicehash stratum mode
"-allpools" option is set, default pools can be used for devfee, check "Readme" file for details.
Watchdog enabled
Remote management is enabled on port 5000

DCR: Stratum - connecting to 'decred.usa.nicehash.com' <198.23.94.139> port 3354
ETH: Stratum - Connected (daggerhashimoto.usa.nicehash.com:3353)
DCR: Stratum - Connected (decred.usa.nicehash.com:3354)
ETH: Authorized
Setting DAG epoch #2...
Setting DAG epoch #2 for GPU0
Create GPU buffer for GPU0
GPU0 DAG creation time - 1704 ms
Setting DAG epoch #2 for GPU0 done

YES, YOU ARE MINING DAGGER-HASHIMOTO ALGORITHM--

And, NiceHash is directing your hash data-stream at different clones of the Etherium (ETH) coin. The "buyers" at NiceHash are paying for your hashing power, but they may select which coin to mine at a different pool. The DAG file for these clone-coins is at varying epochs, and will be of different size according to the epoch of the clone-coin being mined.

--scryptr

cpunkt

newbie

Activity: 6

Merit: 0

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

This does not work for me, after changing the kernel my hashrate is dancing between 25 and 29 Sad

you need to fix your clk level. i do this with roc-smi
look here https://github.com/RadeonOpenCompute/ROC-smi

notbatman

legendary

Activity: 2212

Merit: 1038

Anybody get the open source drivers working? I tried to install fglrx on 16.04 with kernel 4.4 but couldn't get it going.

fgarpe

newbie

Activity: 8

Merit: 0

Here is the solution for linux amdgpu blockchain fix:
The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4

#Install new ROCm Kernel:

wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add -
sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list'
sudo apt-get update
sudo apt-get install rocm

#edit grub
sudo nano /etc/default/grub
#2MB fragments for Ellesmere are enabled with a grub option:
GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"

sudo update-grub
sudo reboot

choose new Kernel and its done!

Thx to gstoner from ROCm Github.

Feel free to donate if you happy ;-)

ETH:0xccad3d66b29ae2924df835df1417169ca59943b5
BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf

Greetz

This does not work for me, after changing the kernel my hashrate is dancing between 25 and 29 Sad

Quote from: jackbox on August 29, 2017, 09:35:16 AM

NoHash

newbie

Activity: 34

Merit: 0

Quote from: Claymore on August 27, 2017, 03:19:42 PM

I am on NiceHash and ONLY have Eth/Dag checked

Nicehash will change between any Ethash coin as they see fit. Not under your control. You pick the algo, not the coin. The people that pay Nicehash pick the coin they want to mine.

NiceHash can only mine those algo's that I have checked on my system,

ETH: nicehash stratum mode
"-allpools" option is set, default pools can be used for devfee, check "Readme" file for details.
Watchdog enabled
Remote management is enabled on port 5000

DCR: Stratum - connecting to 'decred.usa.nicehash.com' <198.23.94.139> port 3354
ETH: Stratum - Connected (daggerhashimoto.usa.nicehash.com:3353)
DCR: Stratum - Connected (decred.usa.nicehash.com:3354)
ETH: Authorized
Setting DAG epoch #2...
Setting DAG epoch #2 for GPU0
Create GPU buffer for GPU0
GPU0 DAG creation time - 1704 ms
Setting DAG epoch #2 for GPU0 done

CarlOrff

sr. member

Activity: 826

Merit: 440

Quote from: ZenFr on August 27, 2017, 02:06:05 PM

Hi Claymore : is there a new version of your dual miner in the pipe, ready soon ?
The actual one is very good, but we ever hope better ;-).

I plan to release new version in 3-4 days. I've almost finished ASM mode for ETH+LBC, also try to find some workaround for voltage control.

Great news : a new version of the Claymore's dual miner soon :-).
I will try the ETH + PASCAL dual mining in next days.

Ursul0

sr. member

Activity: 857

Merit: 262

Quote from: noviceminer on August 29, 2017, 03:53:37 AM

Thats robinh00d version which is an unofficial patch that works in Win 7. We are waiting for something more concrete on the official front.

you should not wait anymore. according to the AMD plans the "blockchain beta" is the only official AMD driver they will release with DAG fix included

Ursul0

sr. member

Activity: 857

Merit: 262

Quote from: NFGG on August 29, 2017, 08:35:44 AM

Hey guys,

I got a question can I redirect the blockchain destination somehow to D: Drive ?? my C: drive is an SSD just for windows. I tried redirecting it through my Command Prompt but it still keeps saving all blockchain data to C:.

Thanks for the help

"blockchain destination" sounds cryptic. if your're talking about geth node you can try "data-dir" (just check the correct syntax) param

jackbox

legendary

Activity: 1246

Merit: 1024