Author

Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 290. (Read 6590718 times)

member
Activity: 140
Merit: 10
That sounds like bad Bios timings. It could also be your -dcri setting. I used 6 or 7 for my RX 580's and 5 for RX 570's. But yes, dual mining profitably is dead for most, at least for now.

I tried stock, slightly modified and one click. Rx580s with dcri at 6.  Claymore v11.5 without oldkernels. Too much fluctuation on the old kernels for me.

Anyways back to single mining.  Starting to see some pretty good rx580 prices on eBay but I wish it would drop more.
hero member
Activity: 1036
Merit: 606
With v11.5 I previously dual mined SmartCash until it was no longer profitable and had at most 1-2 ETH incorrect shares per 1000.

https://bitcointalksearch.org/topic/m.32298051

I'm very happy with the rig stability, even increasing my overclock. It's also run for 3+ days with ZERO problems on v11.5.

https://bitcointalksearch.org/topic/m.32771521

When dual mining you also need to lower your overclock.

I still don't know how you were able to get that low. I'm always around 10 incorrect per 1000 no matter how much I underclocked.  Tried different ram timings, drivers, voltages etc.  Moot point now that keccak is dead.

That sounds like bad Bios timings. It could also be your -dcri setting. I used 6 or 7 for my RX 580's and 5 for RX 570's. But yes, dual mining profitably is dead for most, at least for now.
member
Activity: 198
Merit: 10
I need your advice. I want to mine using Claymore's with AMD+Nvidia cards, but I want to setup with nicehash pool and wallet. Is it possibe ? Nicehash opens 3 different clients but speed is lower than using Claymore.

yes its possibly, just use the nicehash wallet and pool.

read From 1st page

nicehash:
   EthDcrMiner64.exe -epool stratum+tcp://daggerhashimoto.eu.nicehash.com:3353 -ewal 1LmMNkiEvjapn5PRY8A9wypcWJveRrRGWr -epsw x -esm 3 -allpools 1 -estale 0

Thanks! I did it and I found that the speed is better with claymore than nicehash miners. I just use nice hash to run stak-cpu together with claymore.
newbie
Activity: 1
Merit: 0
Lets give it a shot! Grin
member
Activity: 140
Merit: 10
With v11.5 I previously dual mined SmartCash until it was no longer profitable and had at most 1-2 ETH incorrect shares per 1000.

https://bitcointalksearch.org/topic/m.32298051

I'm very happy with the rig stability, even increasing my overclock. It's also run for 3+ days with ZERO problems on v11.5.

https://bitcointalksearch.org/topic/m.32771521

When dual mining you also need to lower your overclock.

I still don't know how you were able to get that low. I'm always around 10 incorrect per 1000 no matter how much I underclocked.  Tried different ram timings, drivers, voltages etc.  Moot point now that keccak is dead.
full member
Activity: 152
Merit: 100
For some reason, since the last windows update, i can no longer use the #0 card, it hangs out setting the dag.
Has anybody experienced this same problem?
I can't go back the windows version, so i'll move to linux (already have a lot of linux rigs so it's not a big problem) but i don't have the time right now, so i wanted to have it fully working for a few days more until the migration Tongue

Thanx guys!
newbie
Activity: 34
Merit: 0

Why are you using such high voltages?
With HWINFO I'm looking how much a card draws and decrease the voltages until there's no variation. Most of them fall below 850mV. And I don't alter powlim.

what do you use to undervolt?
in windows or linux
amazingly -cvddc / -mvddc does not work for me in windows
even in linux i.e. simplemining undervolt does not work

i can only use trixx for my sapphire rx 470 to use 825mV

I use Claymore on my rigs. I tried once, for a customer, to use OverdriveNTool (https://forums.guru3d.com/threads/overdriventool-tool-for-amd-gpus.416116/) and it worked instead of Claymore.
hero member
Activity: 751
Merit: 517
Fail to plan, and you plan to fail.
Im facing a curious issue Claymore. I am running 13 x RX 460 cards on an Asus B250 MiningExpert under Windows 10 (1709).
If I try to run the Miner with all 13 GPU's it just BSOD's right after it lists out all GPU's and starts showing ASM mode for them.
If however, I run two instances of the miner and split the GPU's into groups of 6 and 7 using -di, the two instances run perfectly fine.
I thought you had already optimized the miner for 9+ GPU's? Any Idea why this could be happening?
While two instances of the miner work fine for me, its sub optimal as a solution.

I have a rig based on H110 Pro BTC+ motherboard with 12 GPUs, it works fine (v11.5, latest drivers, Windows 10). Miner itself cannot cause BSOD, it must be something related to PSU (try "-gser 2"), hardware or drivers.
If you think that the reason is the miner, try some other miners to confirm it.


Hmmm. lidag or gser do not help, it just bsod's with atikdmg dll error. Swapped out the PUS's and no change. Additionally, all 13 work fine in one instance when clocks/volts are at stock, but fail when overclocked/undervolted.
What is curious is, I setup a second system exactly the same way today and that works fine overclocked/undervolted with all 13 in one instance of the miner. So now I have no idea why the first system wont.
sr. member
Activity: 1008
Merit: 297
Grow with community
I need your advice. I want to mine using Claymore's with AMD+Nvidia cards, but I want to setup with nicehash pool and wallet. Is it possibe ? Nicehash opens 3 different clients but speed is lower than using Claymore.

yes its possibly, just use the nicehash wallet and pool.

read From 1st page

nicehash:
   EthDcrMiner64.exe -epool stratum+tcp://daggerhashimoto.eu.nicehash.com:3353 -ewal 1LmMNkiEvjapn5PRY8A9wypcWJveRrRGWr -epsw x -esm 3 -allpools 1 -estale 0
newbie
Activity: 11
Merit: 0
Guys,

What is causing this? Everyday almost at the same time on multiple machines...? Anyone with similar issue? Running v11.5....



Quote
14:04:17:909   2324   sent: {"id":6,"jsonrpc":"2.0","method":"eth_submitHashrate","params":["0xcfcef1e", "0x000000000000000000000000000000000000000000000000000000002865ab44"]}

14:04:18:112   2324   ETH: checking pool connection...
14:04:18:112   2324   sent: {"worker": "", "jsonrpc": "2.0", "params": [], "id": 3, "method": "eth_getWork"}

14:04:18:159   2324   buf: {"jsonrpc":"2.0","id":0,"result":["0xd1f9990f4ac5272a35c0e1e7b7209eda946690214df2d93b7d4bf7c82201e0ad","0x3c08da512cf85dc7dd15483f1ebf529b55eb2c4ebc2182a808eebe109eab0e5c","0x000000006df37f675ef6eadf5ab9a2072d44268d97df837e6748956e5c6c2116"]}

                                                                                                                                                                                                                                                                                                                               
newbie
Activity: 15
Merit: 0
Probably a dumb question, but, where do I find the Readme? Huh
On the first post of this topic or in the miner folder there is a file named "Readme!!!"
newbie
Activity: 1
Merit: 0
Probably a dumb question, but, where do I find the Readme? Huh
newbie
Activity: 197
Merit: 0

Why are you using such high voltages?
With HWINFO I'm looking how much a card draws and decrease the voltages until there's no variation. Most of them fall below 850mV. And I don't alter powlim.

what do you use to undervolt?
in windows or linux
amazingly -cvddc / -mvddc does not work for me in windows
even in linux i.e. simplemining undervolt does not work

i can only use trixx for my sapphire rx 470 to use 825mV
donator
Activity: 1610
Merit: 1325
Miners developer
Im facing a curious issue Claymore. I am running 13 x RX 460 cards on an Asus B250 MiningExpert under Windows 10 (1709).
If I try to run the Miner with all 13 GPU's it just BSOD's right after it lists out all GPU's and starts showing ASM mode for them.
If however, I run two instances of the miner and split the GPU's into groups of 6 and 7 using -di, the two instances run perfectly fine.
I thought you had already optimized the miner for 9+ GPU's? Any Idea why this could be happening?
While two instances of the miner work fine for me, its sub optimal as a solution.

I have a rig based on H110 Pro BTC+ motherboard with 12 GPUs, it works fine (v11.5, latest drivers, Windows 10). Miner itself cannot cause BSOD, it must be something related to PSU (try "-gser 2"), hardware or drivers.
If you think that the reason is the miner, try some other miners to confirm it.
newbie
Activity: 4
Merit: 0
Hey guys, have any of you had an issue with V11.5, where when connection was lost to the pool, the miner resumed when the connection was restored, but at a lower hashrate?

At ~1:45am this morning, my miner lost connection, and when the connection was restored, the AMD cards mined at less than 50% of their usual hash rate. Strangely though, the NVidia cards on the rig resumed at their normal hashrate.

AMD Cards: Powercolor RX470, MSI Armor RX480
NVidia Cards: 2 x GTX 1050Ti

The log file:

01:43:23:567   15b0   ETH - Total Speed: 86.369 Mh/s, Total Shares: 913, Rejected: 1, Time: 12:03
01:43:23:567   15b0   ETH: GPU0 28.862 Mh/s, GPU1 29.061 Mh/s, GPU2 14.241 Mh/s, GPU3 14.206 Mh/s

01:45:36:676   15b0   ETH: GPU0 12.901 Mh/s, GPU1 15.053 Mh/s, GPU2 14.250 Mh/s, GPU3 14.232 Mh/s

this happens to me as well for hardcoded bios.

temporary solution is disable p0-p6 in overdriventool. or hardcode your bios to use the same clocks on p0-p6.

this must be a bug, since this is not happening on previous versions.

If you're having this issue on Windows (10), Windows has probably updated the Video Card Driver with his own new driver.  Solution: remove this driver and install AMD's Mining drivers again.  This solved my problem.
newbie
Activity: 37
Merit: 0
Need some help. Besides the 580s I have a gaming pc with a 1080ti in it.
Lately I was using claymore eth miner.
Suddenly I get no hasrate and getting this message

GPU 0, GpuMiner cu_kd failed 77 (0), an illegal memory access was encountered
GPU 0, Calc DAG failed!
GPU 0 failed
Setting DAG epoch #177 for GPU0
GPU 0, CUDA error 77 - cannot write buffer for DAG


On ccminer I get> GPU #0: result for 00000000 does not validate on CPU!


I used MSI afterburner for several weeks without problem. The memory clock was higher I admit, maybe that triggered a lock?
Anyway I am now reverting to older drivers also I am using the default values on the GTX without OC or undervolt. nothing helps.
Tried setting the virtual memory but that was fine, nothing changed.

Any thought?



I will try today to swap the OS (just to change the SSD from the mining rig).
If that helps than it's some driver/OS issue.. I will reinstall Win10, if not I have one more shot with changing the PSU, after that I am pretty clueless.. games run fine though..

I don't get any supported clocks:

C:\Program Files\NVIDIA Corporation\NVSMI>nvidia-smi.exe -q -d PERFORMANCE

==============NVSMI LOG==============

Timestamp                           : Wed Mar 28 08:35:34 2018
Driver Version                      : 391.35

Attached GPUs                       : 1
GPU 00000000:01:00.0
    Performance State               : P0
    Clocks Throttle Reasons
        Idle                        : Not Active
        Applications Clocks Setting : Not Active
        SW Power Cap                : Not Active
        HW Slowdown                 : Not Active
            HW Thermal Slowdown     : Not Active
            HW Power Brake Slowdown : Not Active
        Sync Boost                  : Not Active
        SW Thermal Slowdown         : Not Active
        Display Clock Setting       : Not Active


C:\Program Files\NVIDIA Corporation\NVSMI>nvidia-smi.exe -q -d SUPPORTED_CLOCKS|more

==============NVSMI LOG==============

Timestamp                           : Wed Mar 28 08:36:01 2018
Driver Version                      : 391.35

Attached GPUs                       : 1
GPU 00000000:01:00.0
    Supported Clocks                : N/A
member
Activity: 198
Merit: 10
I need your advice. I want to mine using Claymore's with AMD+Nvidia cards, but I want to setup with nicehash pool and wallet. Is it possibe ? Nicehash opens 3 different clients but speed is lower than using Claymore.
newbie
Activity: 37
Merit: 0
Need some help. Besides the 580s I have a gaming pc with a 1080ti in it.
Lately I was using claymore eth miner.
Suddenly I get no hasrate and getting this message

GPU 0, GpuMiner cu_kd failed 77 (0), an illegal memory access was encountered
GPU 0, Calc DAG failed!
GPU 0 failed
Setting DAG epoch #177 for GPU0
GPU 0, CUDA error 77 - cannot write buffer for DAG


On ccminer I get> GPU #0: result for 00000000 does not validate on CPU!


I used MSI afterburner for several weeks without problem. The memory clock was higher I admit, maybe that triggered a lock?
Anyway I am now reverting to older drivers also I am using the default values on the GTX without OC or undervolt. nothing helps.
Tried setting the virtual memory but that was fine, nothing changed.

Any thought?



I will try today to swap the OS (just to change the SSD from the mining rig).
If that helps than it's some driver/OS issue.. I will reinstall Win10, if not I have one more shot with changing the PSU, after that I am pretty clueless.. games run fine though..
full member
Activity: 298
Merit: 100
i have 7 AMD and 1 Nvidia
Then I don't understand why you mention that you running latest amd drivers in your first post.
Anyway, do you have any errors with your nvidia card in claymore? Do you have -tt or -fanmin/-fanmax set? If yes, set it only for amd cards (from readme: for NVIDIA cards only temperature monitoring is supported, temperature management is not supported.), if no - more info is needed.

issue persist.
22:45:10:453   271c   NVML: cannot get current temperature, error 999 (an internal driver error occurred)
22:45:10:453   271c   NVML: cannot get fan speed, error 999 (an internal driver error occurred)
22:45:13:594   271c   NVML: cannot get current temperature, error 999 (an internal driver error occurred)
22:45:13:594   271c   NVML: cannot get fan speed, error 999 (an internal driver error occurred)

i have a full nvidia build, all 8 cards are 1060s... runs like a champ no issues at all, all my machines are running the latest drivers.

-fanmin 50 -fanmax 80 -tt -50 this is what i have in my line.

how do u set for amd only?
hero member
Activity: 1036
Merit: 606
With v11.5 I previously dual mined SmartCash until it was no longer profitable and had at most 1-2 ETH incorrect shares per 1000.

https://bitcointalksearch.org/topic/m.32298051

I'm very happy with the rig stability, even increasing my overclock. It's also run for 3+ days with ZERO problems on v11.5.

https://bitcointalksearch.org/topic/m.32771521

When dual mining you also need to lower your overclock.
Jump to: