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

[batsonxl]

Hello

Trying to configure strap and rxboost in a rig with 12 gpus I have some doubt, I would appreciate any help

Rig with rx 580, 570 4gb, 8gb, 9 gpus without problem, but in 3 of them I can't apply a strap because they restart.

The 3 gpus do not seem to have anything different from the rest, hynix memory, 4gb, sapphire (2 nitro and 1 pulse).

When I put for example strap 1, they restart, so I have 9 gpus with strap and 3 with strap 0, and rxboost 100 for all.

I have tried using other raisers, connecting a monitor emulator and nothing changes.

I am applying a strap over the previously modified bios.

Maybe if I put original bios strap works?

Apologies for writing bad english.

Regards : )

i wouldnt apply strap to allready modified bios, you have choice use modified bios or stock bios with claymores strap.

flash your cards to stock bios then try again,and also dont forget virtual memory set high as you can.

[1714771218]

1714771218

[batsonxl]

Hi dear miners

is there man who can share stable working setups for these headache memory straps???
i have all samsung elpida micron hynix.... rx470   480   570   580 4...8gb

My some cards runs 2-3 days some of then crash. whatchdog says restart miner.i wish could be nice if it crashes right away, so i know it crashed.it is annoying it crashes and you dont know until you check it.

my rigs are not stable under 930mv on core, 900mv works for 2-3 days some cards just crash after 6 hour.i dont understand why it doesnt crash right away.it is difficult monitor each cards behavior because different model memory brand.and i dont understand people who can get stable card with 835mv??? non of my cards work at this voltage.for me voltage not problem i can waste extra watts for me is important to be stable with NO crashes  at least for a week,then i can restart whole rig

Thanks 

[0xcosmos]

Hi dear miners

is there man who can share stable working setups for these headache memory straps???
i have all samsung elpida micron hynix.... rx470   480   570   580 4...8gb

My some cards runs 2-3 days some of then crash. whatchdog says restart miner.i wish could be nice if it crashes right away, so i know it crashed.it is annoying it crashes and you dont know until you check it.

my rigs are not stable under 930mv on core, 900mv works for 2-3 days some cards just crash after 6 hour.i dont understand why it doesnt crash right away.it is difficult monitor each cards behavior because different model memory brand.and i dont understand people who can get stable card with 835mv??? non of my cards work at this voltage.for me voltage not problem i can waste extra watts for me is important to be stable with NO crashes  at least for a week,then i can restart whole rig

Thanks 

each and every card is different even if you have two identical gpus
it mostly relies on chip quality which can also be checked by gpuz
no two identical cards may run with exactly same configuration
so i will say check the card one at a time for its stable settings

[batsonxl]

Hi dear miners

is there man who can share stable working setups for these headache memory straps???
i have all samsung elpida micron hynix.... rx470   480   570   580 4...8gb

My some cards runs 2-3 days some of then crash. whatchdog says restart miner.i wish could be nice if it crashes right away, so i know it crashed.it is annoying it crashes and you dont know until you check it.

my rigs are not stable under 930mv on core, 900mv works for 2-3 days some cards just crash after 6 hour.i dont understand why it doesnt crash right away.it is difficult monitor each cards behavior because different model memory brand.and i dont understand people who can get stable card with 835mv??? non of my cards work at this voltage.for me voltage not problem i can waste extra watts for me is important to be stable with NO crashes  at least for a week,then i can restart whole rig

Thanks 

each and every card is different even if you have two identical gpus
it mostly relies on chip quality which can also be checked by gpuz
no two identical cards may run with exactly same configuration
so i will say check the card one at a time for its stable settings

Thanks for answer
Check this out : i have 3 rx470 msi armor identical 3 cards memory everything looks like they been manufactured at same time but 2 of them works same setting but 1 insist on 2040mhz memory and 930mv on core,it makes me crazy

[RAKOT]

Hi dear miners

is there man who can share stable working setups for these headache memory straps???
i have all samsung elpida micron hynix.... rx470   480   570   580 4...8gb

My some cards runs 2-3 days some of then crash. whatchdog says restart miner.i wish could be nice if it crashes right away, so i know it crashed.it is annoying it crashes and you dont know until you check it.

my rigs are not stable under 930mv on core, 900mv works for 2-3 days some cards just crash after 6 hour.i dont understand why it doesnt crash right away.it is difficult monitor each cards behavior because different model memory brand.and i dont understand people who can get stable card with 835mv??? non of my cards work at this voltage.for me voltage not problem i can waste extra watts for me is important to be stable with NO crashes  at least for a week,then i can restart whole rig

Thanks 

each and every card is different even if you have two identical gpus
it mostly relies on chip quality which can also be checked by gpuz
no two identical cards may run with exactly same configuration
so i will say check the card one at a time for its stable settings

Thanks for answer
Check this out : i have 3 rx470 msi armor identical 3 cards memory everything looks like they been manufactured at same time but 2 of them works same setting but 1 insist on 2040mhz memory and 930mv on core,it makes me crazy

Agree, GPUs doesnt made by copy-paste method

[daworm06]

I've updated to Claymore V15. It says 16 pools are set for this miner but it is mining on my account at nano pool.  How do I delete the other pools?

[Bael]

Hello

I have an h110 pro btc motherboard, and after going around many times because 3 gpus could not apply -strap I discovered that you have to do with pcie ports.

I explain, for some reason 3 of the 13 ports if I connect a gpu to them does not allow to apply -strap

I have checked changing ports gpus, risers, cables, etc. and it is always those three ports that do not allow to apply -strap, if I put in those 3 ports -strap 0 that gpus work well, but with its modified bios (31 mhs) or bios stock (25 mhs)

I thought it could be the bios gen configuration, so I tried to change gen1, gen2 and automatic gen, but without changes.

Apparently those 3 pcie ports are not broken, using bios mods the gpus reach 31 mhs, normal would say, but as soon as I apply -strap they restart.

Anyone know how to shed light on the problem? : )

thanks!!


edit: rig with 12 gpus, rx 580, 570 of 4 and 8 GB

[tg88]

Hello

I have an h110 pro btc motherboard, and after going around many times because 3 gpus could not apply -strap I discovered that you have to do with pcie ports.

Is the motherboard BIOS updated?

Did you connect the main power supply to the PCIE PWR and SATA PWR connectors?

[Jonet]

i wouldnt apply strap to allready modified bios, you have choice use modified bios or stock bios with claymores strap.

flash your cards to stock bios then try again,and also dont forget virtual memory set high as you can.

Claymore -strap will override whatever memory timings are set in BIOS. It does not matter if you have stock timing or have flashed a modified BIOS. Restoring to original BIOS is nice if you sometimes use the GPU for gaming or want to have the card ready for sale, but makes no difference for mining when you use the -strap option in Claymore.

[Bael]

Hello

I have an h110 pro btc motherboard, and after going around many times because 3 gpus could not apply -strap I discovered that you have to do with pcie ports.

Is the motherboard BIOS updated?

Did you connect the main power supply to the PCIE PWR and SATA PWR connectors?

Hello

Thanks for the help!!

I have the 3 connectors to the main power supply and I use the latest bios version of the motherborad, the 1.60

What I don't understand is because the ports work well without applying -strap, maybe they need more power?

[Bael]

i wouldnt apply strap to allready modified bios, you have choice use modified bios or stock bios with claymores strap.

flash your cards to stock bios then try again,and also dont forget virtual memory set high as you can.

Claymore -strap will override whatever memory timings are set in BIOS. It does not matter if you have stock timing or have flashed a modified BIOS. Restoring to original BIOS is nice if you sometimes use the GPU for gaming or want to have the card ready for sale, but makes no difference for mining when you use the -strap option in Claymore.

True, proven that this is so, I have done several tests lately, no matter how to use stock bios or modified bios, -strap applies the same.

[Jonet]

If I use the latest version, the 1.60 integrated port of the gpu does not work for me, I don't know what is the reason for this.

Issues with Windows 10 - Intel Integrated Graphics - Claymore can be solved by installing the latest graphics driver from Intel.com (Not windows update).
I ran into a problem a while ago where claymore would crash if no monitor was attached to the motherboard. May not be the same problem but could be worth it to try.
Apart from that I don't understand why you need to use integrated graphics on a mining rig unless you are using those purpose built mining GPUs with no connectors.

[neca1]

If I use the latest version, the 1.60 integrated port of the gpu does not work for me, I don't know what is the reason for this.

Issues with Windows 10 - Intel Integrated Graphics - Claymore can be solved by installing the latest graphics driver from Intel.com (Not windows update).
I ran into a problem a while ago where claymore would crash if no monitor was attached to the motherboard. May not be the same problem but could be worth it to try.
Apart from that I don't understand why you need to use integrated graphics on a mining rig unless you are using those purpose built mining GPUs with no connectors.

hello help me please what is it?  eth: Authorization failed : {"id":2,"jsonrpc":"2.0","result":null,"error":{"code":-1,"message":"Youareblacklisted,pleasecontacthelpdeskwithyourdetails"}} DevFee: ETH: Job timeout, disconnect, retry in 20 sec...

I have win10 adrenalin 2019, driver 19.9.2 GPU VEGA AMD 56 8GB memory.


Thi is claymore dual AMD  EthDcrMiner64.exe -epool stratum+tcp://eu.anyhash.net:8018 -ewal 0x7d64450ad2c4bfdXXXXXXXXXXXXXXXXXXXXXXXXX -epsw x -rxboost 1

Thank for ansver

[delevic]

Hi Claymore,

I have little problem in Ubuntu. When I start script with oc and underwalt command before I start miner , miner turn off gpu1 always. When I first start miner  then script everything is ok. What it can be?

[Bael]

If I use the latest version, the 1.60 integrated port of the gpu does not work for me, I don't know what is the reason for this.

Issues with Windows 10 - Intel Integrated Graphics - Claymore can be solved by installing the latest graphics driver from Intel.com (Not windows update).
I ran into a problem a while ago where claymore would crash if no monitor was attached to the motherboard. May not be the same problem but could be worth it to try.
Apart from that I don't understand why you need to use integrated graphics on a mining rig unless you are using those purpose built mining GPUs with no connectors.

Yes, it was solved by updating the intel graphic driver, thanks!! : ), now I have the bios 1.60

I use the integrated one to not overload any gpu when I connect remotely, actually there I have read that 14 gpus are connected on this motherboard, logically deactivating the integrated one and using a pcie multiplier card.

From my personal experience if I put 13 gpus (plus the integrated one) it starts to give incorrect share errors in some random gpu, for me the maximum reliable is 12.

Regarding my problem with the pcie ports it remains the same, 3 of the 13 connectors do not support -strap, so I have 10 working with -strap and 2 with mod bios

[0xcosmos]

Hi Claymore,

I have little problem in Ubuntu. When I start script with oc and underwalt command before I start miner , miner turn off gpu1 always. When I first start miner  then script everything is ok. What it can be?

the script to oc and undervolt your card is interacting with which software? can you post the script code?

[delevic]

Hi Claymore,

I have little problem in Ubuntu. When I start script with oc and underwalt command before I start miner , miner turn off gpu1 always. When I first start miner  then script everything is ok. What it can be?

the script to oc and undervolt your card is interacting with which software? can you post the script code?

With Claymore miner somehow. If I executed script first, clay turns off gpu1.

Code:

#!/bin/bash

sudo  nvidia-smi -pm 1 && sudo nvidia-smi -pl 80
nvidia-settings -a '[gpu:0]/GPUMemoryTransferRateOffset[2]=800'
nvidia-settings -a '[gpu:1]/GPUMemoryTransferRateOffset[2]=800'
nvidia-settings -a '[gpu:2]/GPUMemoryTransferRateOffset[2]=800'
nvidia-settings -a '[gpu:3]/GPUMemoryTransferRateOffset[2]=800'
nvidia-settings -a '[gpu:4]/GPUMemoryTransferRateOffset[2]=800'

And I tested ethminer , gpu works fine.

[patinencomun]

Hi, I bought a 5700, seems there is no optimizations anywhere, I got 51MH at 110w, not bad, but sure can do 60+

Anyone knows any bios editor for 5700 series?
Claymore are you working on 5700 series?
Thank you

[adlian]

Hi there,

after update to Win 1909 there is problem with the drivers.

I use Win10-64Bit-Crimson-ReLive-Beta-Blockchain-Workloads-Aug23 (AMD Radeon RX480 and 470 mix)  and it work for a day , than it crashes
without any update notice : red open gl problem ;

I used amd clener and installed the last AMD driver from 2.12.19 - but the miner don't work ;

than again cleaner and again the blockchain drv - it works for a day-two ... and crash again ?

Pls help

[abg00]

Hi there,

after update to Win 1909 there is problem with the drivers.

I use Win10-64Bit-Crimson-ReLive-Beta-Blockchain-Workloads-Aug23 (AMD Radeon RX480 and 470 mix)  and it work for a day , than it crashes
without any update notice : red open gl problem ;

I used amd clener and installed the last AMD driver from 2.12.19 - but the miner don't work ;

than again cleaner and again the blockchain drv - it works for a day-two ... and crash again ?

Pls help

check -rxboost 2 if use -rxboost 1 now