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

[bluspirit]

version 15 works correctly up to the 376 epoch. The new version is not needed yet

when using Windows the first GPU takes something like 140-150 MB more video memory on top of the other ones that are just mining, so with that the actually used video memory gets closer to the 4GB limit and instead of 20-30 MH/s you might be getting just 1-2 MH/s for the first GPU and soon it might even stop working.

First GPU always use more video memory for system stuff.
So the main problem does not depend on CM version but on size of the DAG file required to mine etherium and loaded into the GPU memory. The DAG size increases by 8MB in a few days when the epoch changes. Therefore it's inevitable that soon 4GB GPUs will no longer be able to mine ETH, as happened in the past years for the 2GB / 3GB GPUs.

In short term some workarounds could be:
1) Use a different PCI slot (different from first: PCIx16_1) for 4GB GPUs or enable integrated GPU.
2) Switch to linux which uses less memory for first GPU.

Personally, although I am still able to mine ETH with CM 15 / WIN10 even with a 4GB GPU in first PCI slot, sometimes when the epoch changes the hashrate decreases, I have to manually restart CM or Windows. In my case, other miners no longer work for 4GB GPU: they already start working with a low hashrate.

[RAKOT]

version 15 works correctly up to the 376 epoch. The new version is not needed yet

when using Windows the first GPU takes something like 140-150 MB more video memory on top of the other ones that are just mining, so with that the actually used video memory gets closer to the 4GB limit and instead of 20-30 MH/s you might be getting just 1-2 MH/s for the first GPU and soon it might even stop working.

First GPU always use more video memory for system stuff.
So the main problem does not depend on CM version but on size of the DAG file required to mine etherium and loaded into the GPU memory. The DAG size increases by 8MB in a few days when the epoch changes. Therefore it's inevitable that soon 4GB GPUs will no longer be able to mine ETH, as happened in the past years for the 2GB / 3GB GPUs.

In short term some workarounds could be:
1) Use a different PCI slot (different from first: PCIx16_1) for 4GB GPUs or enable integrated GPU.
2) Switch to linux which uses less memory for first GPU.

Personally, although I am still able to mine ETH with CM 15 / WIN10 even with a 4GB GPU in first PCI slot, sometimes when the epoch changes the hashrate decreases, I have to manually restart CM or Windows. In my case, other miners no longer work for 4GB GPU: they already start working with a low hashrate.

1) Use a different PCI slot (different from first: PCIx16_1) for 4GB GPUs or enable integrated GPU.

IMHO there is no point in that. Any GPU with attached monitor will be the first GPU and will display dektop , eating memory from GPU, so it will have less RAM to hold DAG file and we will face this issue.

I think of 3 ways : 1. GPU0 with 6-8 Gb RAM. 2. Inegreted GPU with attached monitor. 3. Linux distro as it consumes less RAM ( will do but will be the day it fail)

[iDisturbia]

This dual mining for supported coins -> Decred/Siacoin/Lbry/Pascal/Blake2s/Keccak

Anyone mine them? Are they even worth it? whattomine doesn’t list them under their GPUs tab!

I just want to make sure their developers have not opened up to ASIC only!!

[br2459]

This dual mining for supported coins -> Decred/Siacoin/Lbry/Pascal/Blake2s/Keccak

Anyone mine them? Are they even worth it? whattomine doesn’t list them under their GPUs tab!

I just want to make sure their developers have not opened up to ASIC only!!

Siacoin is definitely ASIC only, I had a friend that had 7 ASICs for Siacoin until they forked and made them worthless. Not sure about the rest. I don't know if anyone dual mines anymore, I always stayed straight ETH.

[itsmesri123]

I am RIG has 6 - RX 570 cards. If I use Claymore V15. I am getting DAG error. Please tell how to resolve this. All 6 card  Sapphire 11266-04-20G Radeon PULSE RX 570 4GB GDDR5 DUAL HDMI / DVI-D /

Code:

ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU0, OpenCL error -4 - cannot write buffer for DAG
GPU 0 failed
Setting DAG epoch #328 for GPU0
ETH: 04/10/20-00:11:40 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:00
ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU 3, GpuMiner cu_kd2b failed, not enough memory
GPU0, OpenCL error -4 - cannot write buffer for DAG
GPU 5, GpuMiner cu_kd2b failed, not enough memory
GPU 1, GpuMiner cu_kd2b failed, not enough memory
ETH: 04/10/20-00:11:44 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:00
ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU 3 failed
GPU 5 failed
GPU 0 failed

https://i.stack.imgur.com/1bPH9.png

[sanhwy]

You have error OpenCL error -4
You need to increase your virtual memory from
control panel - system - advanced system settings - advanced - settings - advanged - change

I am RIG has 6 - RX 570 cards. If I use Claymore V15. I am getting DAG error. Please tell how to resolve this. All 6 card  Sapphire 11266-04-20G Radeon PULSE RX 570 4GB GDDR5 DUAL HDMI / DVI-D /

Code:

ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU0, OpenCL error -4 - cannot write buffer for DAG
GPU 0 failed
Setting DAG epoch #328 for GPU0
ETH: 04/10/20-00:11:40 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:00
ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU 3, GpuMiner cu_kd2b failed, not enough memory
GPU0, OpenCL error -4 - cannot write buffer for DAG
GPU 5, GpuMiner cu_kd2b failed, not enough memory
GPU 1, GpuMiner cu_kd2b failed, not enough memory
ETH: 04/10/20-00:11:44 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:00
ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 0.000 Mh/s, GPU5 0.000 Mh/s
GPU 3 failed
GPU 5 failed
GPU 0 failed

https://i.stack.imgur.com/1bPH9.png

[itsmesri123]

I think have a pretty good number on virtual memory.  Initial Size(MB): 3000 Maximum size (MB): 20000


https://i.stack.imgur.com/pjEVG.png

[iDisturbia]

Siacoin is definitely ASIC only, I had a friend that had 7 ASICs for Siacoin until they forked and made them worthless. Not sure about the rest. I don't know if anyone dual mines anymore, I always stayed straight ETH.

Thank YOU for your time and info.

I’ve been doing the same - just ETH but became curious after reading some recent articles from certain online publishers suggesting dual-mining is still a thing and referring to Claymore as best and only option!

Stay safe.

[big_daddy]

I think have a pretty good number on virtual memory.  Initial Size(MB): 3000 Maximum size (MB): 20000

no is not enough
raise it to 48.000 it should solve your problem

switch your mining to a linux distribution cause your cards of 4gb will mine much longer on linux than on windows, and linux is more stable

[Dowper_]

You cannot mine on 4GB GPU if it's on the first slot (main GPU), you need to enable integrated GPU or to install 8GB as main GPU.

[VictorGT]

version 15 works correctly up to the 376 epoch. The new version is not needed yet

when using Windows the first GPU takes something like 140-150 MB more video memory on top of the other ones that are just mining, so with that the actually used video memory gets closer to the 4GB limit and instead of 20-30 MH/s you might be getting just 1-2 MH/s for the first GPU and soon it might even stop working.

If you use the mobo IGPU for monitor, difference between first slot GPU and others is about 80 MB (Windows 10, GPU-Z).

[RAKOT]

I think have a pretty good number on virtual memory.  Initial Size(MB): 3000 Maximum size (MB): 20000

You should set the amout of VMEM like summ of all GPUs RAM size.

[Dowper_]

Phoenix miner vs Claymore miner comparison - https://www.youtube.com/watch?v=qzbwVkW5QyI

[RAKOT]

Hello!

Anyone tested the mixed setup ( nvidia + radeon) ? Any tips of setting this right ?

[bategojko74]

Phoenix miner vs Claymore miner comparison - https://www.youtube.com/watch?v=qzbwVkW5QyI

Dowper, you are just losing your time. Everyone here will label you a F....N TROLL, just because they are blind to see the simple truth. They don't want it. Here the truth has no value. These people have earned much more with the legendary Claymore miner when the Ethereum was $1600 USD so why should they bother to change their miner just for 4% more of $160 USD per 1 Ethereum. They have emotional affection with Claymore. Emotions are stronger than logic in most people.

[issie81]

Hey guys,

i am old school miner on Vega 56:)

i used TRM doing 44.5Mh but now i want to use Claymore (it's more legend miner), currently doing 43.5Mh losing 1Mh a card, i am mining with 380x Vega 56 and using MmpOS as OS is there maybe a switch optimizing the hashrate 1Mh loss might not seem alot but multiple with 380x it's actually alot

Cheers!

PS i use core 950 / mem 940Mhz (Hynix) or 960Mhz (Samsung) @ 837

[impynick]

Hey guys,

i am old school miner on Vega 56:)

i used TRM doing 44.5Mh but now i want to use Claymore (it's more legend miner), currently doing 43.5Mh losing 1Mh a card, i am mining with 380x Vega 56 and using MmpOS as OS is there maybe a switch optimizing the hashrate 1Mh loss might not seem alot but multiple with 380x it's actually alot

Cheers!

PS i use core 950 / mem 940Mhz (Hynix) or 960Mhz (Samsung) @ 837

TRM gets 50Mh on hynix vega56. You're leaving more than 10% on the table by switching

[sxemini]

I think have a pretty good number on virtual memory.  Initial Size(MB): 3000 Maximum size (MB): 20000

no is not enough
raise it to 48.000 it should solve your problem

switch your mining to a linux distribution cause your cards of 4gb will mine much longer on linux than on windows, and linux is more stable

Linux more stable? Why? Explain!!!!!!!!!!!!!!!
I think you are wrong, setup win10 in the right way, it is also stable. Running my Rigs on win10 over 3 or 4 months without reboot is no problem.

[big_daddy]

Linux more stable? Why? Explain!!!!!!!!!!!!!!!
I think you are wrong, setup win10 in the right way, it is also stable. Running my Rigs on win10 over 3 or 4 months without reboot is no problem.

I had the complete different suituation, mining on W7, later on W10 gave me such problems with stability and also hashrate drops that I finally decide to move on linux completely
Linux cannot freez, that's the main point, than I can still mine with 4GB cards

the remote management is much more easier too

[waywoos]

i have a simple question for someone im sure.

i almost always us phoenix miner.

i want to know if there is a command to show me how many shares each of the cards is getting.
claymore only gives me a total and doesn't break it down like phoenix does.
and as well what is the command to stop it from writing a log file after every session?
thanks in advance