Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux)

[fgarpe]

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

https://s26.postimg.org/t740rxkmx/rocm.png

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

I find the problem, I was using ohgodatool to overclock memory

Code:

ohgodatool -i 0 --mem-state 2 --mem-clock 2150

Without overclocking memory I get a stable hasrate in 27.8 aprox, is there any way to overclock memory with rocm utilities?

Regards

[1714864383]

1714864383

[fgarpe]

Hi,

I just need to verify something about the amd beta blockchain driver.
I have a rig 8 GPU RX580 Gaming 8 Go (Hynix), in june i got 242 MH/s.
Actually in august i obtain 233 MH/s, in order to solve this drop MH/s, i installed this beta blockchain driver i got 248 MH/s, but i met issue with GPU clock speed frequency alternatively jumping 1000 to 1300 MHz.
Is somebody here met the same issue and find solution to solve this?

Thank you for help.

Had the same problem with my RX570, also befor kernel update. I flashed only 1 clk level to the RX570, after that they run fine.
With the RX4X0 i had no Problems to set the clk with rocm-smi, they run stabel

I have no issue with official driver, i only meet this issue gpu clock frequency jumping alternatively non stop from 1000 to 1300 MHz with the beta blockchain driver.
I dont see what you mean, you flashed only 1 clk level to solve this issue?

Can you undervolt with the blockchain driver?

Yes, but only with Afterburner version beta 4.4.0.16, i found it on the link below, i tried to undervolt it was ok for consumption, but jumping issue always there.
https://forums.guru3d.com/threads/rtss-6-7-0-beta-1.412822/page-32

You don't need flash to obtain only ine clk level, you can do it with rocm utilities. Do that before executing Claymore:

Code:

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

Regards

[cpunkt]

Hi,

I just need to verify something about the amd beta blockchain driver.
I have a rig 8 GPU RX580 Gaming 8 Go (Hynix), in june i got 242 MH/s.
Actually in august i obtain 233 MH/s, in order to solve this drop MH/s, i installed this beta blockchain driver i got 248 MH/s, but i met issue with GPU clock speed frequency alternatively jumping 1000 to 1300 MHz.
Is somebody here met the same issue and find solution to solve this?

Thank you for help.

Had the same problem with my RX570, also befor kernel update. I flashed only 1 clk level to the RX570, after that they run fine.
With the RX4X0 i had no Problems to set the clk with rocm-smi, they run stabel

I have no issue with official driver, i only meet this issue gpu clock frequency jumping alternatively non stop from 1000 to 1300 MHz with the beta blockchain driver.
I dont see what you mean, you flashed only 1 clk level to solve this issue?

Can you undervolt with the blockchain driver?

Yes, but only with Afterburner version beta 4.4.0.16, i found it on the link below, i tried to undervolt it was ok for consumption, but jumping issue always there.
https://forums.guru3d.com/threads/rtss-6-7-0-beta-1.412822/page-32

You don't need flash to obtain only ine clk level, you can do it with rocm utilities. Do that before executing Claymore:

Code:

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

Regards

i do this all direct in bios of RX undervolt and oc sclk and mclk.
try direkt memory oc with this command, this gives u 5% more mclk:

echo 5 > /sys/class/drm/card0/device/pp_mclk_od

maybe it works

[catalinionut229]

Can anyone explain to me why I'm getting this error when mining ubiq?
I mining about 10 minutes later I get this error
 opencl error -30
opencl error -61

not enough GPU memory to place DAG, you cannot mine this coin with this GPU

This is the command and I use 2 gb rx 460 video cards 6x
EthDcrMiner64.exe -epool stratum+tcp://us.ubiqpool.io:8008  -ewal 0xAe291bbD9E08F624e07B7E27Ac0fea479468846D0 -epsw x -eworker rigRx460 -allpools 1 -mode 1

[CrayjeksAjek]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

You upgraded to windows 10, but did you remember to set 16-32GB Pagefile?  Are you using AB 4.4.16 Beta to OC?

Yes using AB 4.4.0 Beta 16
I did forget to check page file after updating. Ill have to see once Im home. I hope its somthing silly like that and it works again >_<

[CrayjeksAjek]

Can anyone explain to me why I'm getting this error when mining ubiq?
I mining about 10 minutes later I get this error
 opencl error -30
opencl error -61

not enough GPU memory to place DAG, you cannot mine this coin with this GPU

This is the command and I use 2 gb rx 460 video cards 6x
EthDcrMiner64.exe -epool stratum+tcp://us.ubiqpool.io:8008  -ewal 0xAe291bbD9E08F624e07B7E27Ac0fea479468846D0 -epsw x -eworker rigRx460 -allpools 1 -mode 1

Get 4gb or 8gb cards

[Yohow]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

You upgraded to windows 10, but did you remember to set 16-32GB Pagefile?  Are you using AB 4.4.16 Beta to OC?

Yes using AB 4.4.0 Beta 16
I did forget to check page file after updating. Ill have to see once Im home. I hope its somthing silly like that and it works again >_<

I have a question: what is better? 16 or 32? With 16 isn't enough? I'm using W10 Pro 64bits, 8 GB RAM, G1840 processor, 6x RX 480s (8GB) and H97 Anniversay mobo. Now I'm using 16 but I can't select 2150 for memmory clock because GPU crashes. For now I put 1150/2100 and stable minus GPU0 which is in 1150/2080. Do you think of I set virtual memmory at 32 fix it? Or simply the GPU isn't able to OC more?

[CrayjeksAjek]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

You upgraded to windows 10, but did you remember to set 16-32GB Pagefile?  Are you using AB 4.4.16 Beta to OC?

Yes using AB 4.4.0 Beta 16
I did forget to check page file after updating. Ill have to see once Im home. I hope its somthing silly like that and it works again >_<

I have a question: what is better? 16 or 32? With 16 isn't enough? I'm using W10 Pro 64bits, 8 GB RAM, G1840 processor, 6x RX 480s (8GB) and H97 Anniversay mobo. Now I'm using 16 but I can't select 2150 for memmory clock because GPU crashes. For now I put 1150/2100 and stable minus GPU0 which is in 1150/2080. Do you think of I set virtual memmory at 32 fix it? Or simply the GPU isn't able to OC more?

Try it and see. Couldnt hurt anything.

[bardacuda]

Can anyone explain to me why I'm getting this error when mining ubiq?
I mining about 10 minutes later I get this error
 opencl error -30
opencl error -61

not enough GPU memory to place DAG, you cannot mine this coin with this GPU

This is the command and I use 2 gb rx 460 video cards 6x
EthDcrMiner64.exe -epool stratum+tcp://us.ubiqpool.io:8008  -ewal 0xAe291bbD9E08F624e07B7E27Ac0fea479468846D0 -epsw x -eworker rigRx460 -allpools 1 -mode 1

It's probably trying to mine Ethereum for the devfee...but it's weird that it happens after only 10 minutes and not 15. If you use -allcoins exp it will mine expanse for the devfee which will be fine for 2GB cards.

[ledio]

Hi.

I noticed a couple of problems with the miner.

1. ethpool.org IP is not resolved correctly. Looks like eu1.ethpool.org IP is hard-coded or cached somewhere into miner. First connection attempt ALWAYS fails. Then, 20 sec wait, then, successful connect using a failover address. The first IP that miner tries to connect to (91.121.73.83) is not the IP which eu1.ethpool.org actually resolves to at the moment (188.165.235.31). My connections string is ./ethdcrminer64 -epool eu1.ethpool.org:3333 (Ubuntu)

2. I mine thru a typical wired router. When sometimes the router looses ISP connection it, as usual, serves its own error page to all clients. The miner, however, thinks it's a kind of a local proxy and attempts to continue mining. As no more new shares are found (because no more new jobs supplied), the miner thinks it's GPU failure ("Miner thread hangs") and activates the watchdog, instead of simply sitting there and waiting for internet connection to resume. This is not the right type of behavior, IMO.

Thanks!

1)
Pinging eu1.ethpool.org [91.121.73.83] with 32 bytes of data:
Reply from 91.121.73.83: bytes=32 time=74ms TTL=56
Reply from 91.121.73.83: bytes=32 time=75ms TTL=56
Reply from 91.121.73.83: bytes=32 time=76ms TTL=56
Reply from 91.121.73.83: bytes=32 time=74ms TTL=56

2) what is that: " serves its own error page to all clients"?   - this is NOT how it should behave, nor claymore miner. sort your issues out, yet better figure out what exactly is happening. meantime you an use "-r 1" option with reboot.bat file to solve the hangs

I have same problem. Workers mined ok and in one moment cant connect any more to any server to ethpool.org . From my location i can ping all servers an it is ok but miner wont connect. I think it is Claymore problem.

[Falcon494]

HI GUYS,

 I need help !!!

CLAYMORE V9.8 mining ethereum

I am using ZOTAC GTX 1070 MINI GPU , with Intel pentium G4400 CPU, 4 GB DDR4 ram, 1500 Watt PSU silver edition

WIN 10 Verision 1703 , NVIDIVA Driver verision 385.41

everything works fine for 5 GPU, and basic hashrate 26 Mh/Sec and after MSI afterburner overclocking 30-31Mh/sec

PROBLEM :

when I add the 6th Card..  it runs for 3-5 mins and restarts windows ... it is like loop .

---
I even tried BIOS motherboard H110-D3A Gigabyte enabling 4G encoding.


still, it crashes and restart.

how to fix this issue..

any help... I even checked PCI riser cables .. unpluged and re plugged..

how to make 6th card working without crashing windows

Sounds like your psu is being overloaded by that 6th card.

I run 7x 1070 ITX stable on 1200W PSUs so 1500W should be heaps. As an aside I'd look at at least a gold quality PSU if you can swing it. Unless you have cheap power.

Check your BIOS version. Looks like it should be F23a which was relased in July https://www.gigabyte.com/Motherboard/GA-H110-D3A-rev-10#support-dl

Install the latest nvidia driver. The advice to use a particular version is out of date.

Don't OC until you get them all running to remove that as a factor.

If it's still resetting try a fresh install of Windows from scratch.

Thanks

But still even after BIOS update of motherboard version F23a , Re-installed drivers .. still it is crashing.. last option I need to re-install windows..

should I use only original windows so it doesnt crash.. and reboot

also checked windows event viewer which shows Kernel-PnP driver not responding and critical kernel power

[CrayjeksAjek]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

You upgraded to windows 10, but did you remember to set 16-32GB Pagefile?  Are you using AB 4.4.16 Beta to OC?

Page file was set to 50GB so that's not the issue; I changed it to 64GB just because.

[TechPark]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

I have 2 rigs with RX580: W10+Aug11 Blockchain drivers=have no issues - works flawlessly for weeks.
I do OC sanely though. Try to run it with default settings.

[TechPark]

Hi,

I just need to verify something about the amd beta blockchain driver.
I have a rig 8 GPU RX580 Gaming 8 Go (Hynix), in june i got 242 MH/s.
Actually in august i obtain 233 MH/s, in order to solve this drop MH/s, i installed this beta blockchain driver i got 248 MH/s, but i met issue with GPU clock speed frequency alternatively jumping 1000 to 1300 MHz.
Is somebody here met the same issue and find solution to solve this?

Thank you for help.

Had the same problem with my RX570, also befor kernel update. I flashed only 1 clk level to the RX570, after that they run fine.
With the RX4X0 i had no Problems to set the clk with rocm-smi, they run stabel

I have no issue with official driver, i only meet this issue gpu clock frequency jumping alternatively non stop from 1000 to 1300 MHz with the beta blockchain driver.
I dont see what you mean, you flashed only 1 clk level to solve this issue?

Can you undervolt with the blockchain driver?

Yes, but only with Afterburner version beta 4.4.0.16, i found it on the link below, i tried to undervolt it was ok for consumption, but jumping issue always there.
https://forums.guru3d.com/threads/rtss-6-7-0-beta-1.412822/page-32

I can report that undervolting via Trixx is working, without the jumping issue.

https://farm5.staticflickr.com/4340/35855311444_e79c23f28c_b.jpg

What version of TRIXX?

[asmodyus]

Hello I am new to trying to mine Ethereum I have the new miner and using GTX 1080 ti on windows 10 using nvidia driver 384.94 trying to mine at supernova and I am getting eth: job timeout, disconnect and Received unknown response: {jsonrpc":"2.0","result';[0x0555531 etc,,,,,

Need some help please.

Anybody?

[P00P135]

Hello I am new to trying to mine Ethereum I have the new miner and using GTX 1080 ti on windows 10 using nvidia driver 384.94 trying to mine at supernova and I am getting eth: job timeout, disconnect and Received unknown response: {jsonrpc":"2.0","result';[0x0555531 etc,,,,,

Need some help please.

Anybody?

Use nanopool

[fpontilho]

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

I'm getting the following error:

============ Starting Miner ===============
dirname: missing operand
Try 'dirname --help' for more information.
/root/xminer.sh: line 40: [: !=: unary operator expected
GO TO GLOBAL OPTIONS or RELOAD RIG if you already configured options and this ris is new added !!!!!


I think I don't know how to chose the kernel after all.. Would you help me out?

Anybody?

[joaocha]

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

I'm getting the following error:

============ Starting Miner ===============
dirname: missing operand
Try 'dirname --help' for more information.
/root/xminer.sh: line 40: [: !=: unary operator expected
GO TO GLOBAL OPTIONS or RELOAD RIG if you already configured options and this ris is new added !!!!!


I think I don't know how to chose the kernel after all.. Would you help me out?

Anybody?

sudo nano  /etc/default/grub

Just edit these 2 lines

GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9"
GRUB_DEFAULT="Advanced options for Ubuntu>Ubuntu, with Linux 4.11.0-kfd-compute-rocm-rel-1.6-148"

sudo update-grub
sudo reboot

[Husherig]

Any one else having problem with Win10 freezing?
I really like getting 117mhs again with my 4 cards but if this driver is just going to give headache after headache Im rolling back to Win7 and keeping my 105mhs.
I lowered mem from 2150 to 2100 and it ran a bit longer (previously 10 minutes before crashed windows) but its still crashing the system. Will new miner address issues with Win7 and blockchain drivers? Can anyone give advice for Win10 crashing with new drivers?

I have 2 rigs with RX580: W10+Aug11 Blockchain drivers=have no issues - works flawlessly for weeks.
I do OC sanely though. Try to run it with default settings.

Hi TechPark, how many GPU's you have on each rig, miner Claymore v9.8 and setting? what motherboard?

[Marvell1]

BE CAREFUL with ETHPOOL

oh finally i was watching my miners and I notice some of my miners will get the message "share rejected by pool"

then they disconnect you you get

or cannot connect to "US1.ethpool.org:3333"

then disconect and the re-connect, and then some of the rigs cards only hash at 8.345 and some rigs even just die.

Its a persistent problem, I will give it a few more days then I', m moving my hash to nanopool , its really fking anoying man

I have seen that before but with the new blockchain drivers rigs are more sensitive to crappy pools and constant disconects