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

[svyatko]

hey guys. what sort of decent hashrate should i be expecting from an Asus dual 1070 card ? i am getting an average ETH @ 26.200Mh/s and DRC @ 435MH/s . Does this sound right ?

I'm getting ETH 30 and SIA 350 with that exact card. Memclock +530.

hmm, interesting indeed.
Is yours OC edition and what OS are you on ?

i am on Win10, using GPU Tweak tool from asus, which doesnt let me do any more than +480 on the memclock, making it @ 8488Mhz. With mem on max i can now get 27.400mhs ETH / 456mhs DCR

[Ursul0]

Hi guys who tried new drivers 17.12.1 (adrenaline), Claymore 10.2 can not control frequencies and fans

Tried them last night but immediately rolled back to previous driver. Looks like none of the config parameters are being passed through to cards. AMD have implemented the ability to save profiles in Wattman so I guess maybe something about that has broken that functionality in Claymore?? Also 17.12.1 reoriented all my cards, not a huge deal but still a pain in the butt with my setup.

In case you have AMD gpus (which is likely the case) you should not have been using claymore freq/volts config from claymore.
do the overdriven tool instead. and the only thing driver gui is needed is for setting "compute" mode for each card.
btw did they by any chance change their "settings" app and maybe like added a single option "set compute mode to all gpus" ? - that would be a a good reason for me to switch from 17.11.1?

[Ursul0]

hi guys,

I just built a new rig, Fatal1ty H270 Performance with 4GB ram, I3 cpu and 6 x asus radeon rx 570 expedition oc

When I installed 2 cards first on their own in PCIe slots, then windows 10 etc , blockchain drivers plsu patchers I would get solid 27MH/s (BIOS moded on all cards).

then, I moved the mobo to open case and plugged all six GPUs in using PCIe risers, installed drivers etc, then used the patcher, I can see in GPUz all cards have correct drivers etc.

I run Claymore , it detects 6 cards, stars mining and all I get is 21mh/s

any idea what am I doing wrong please?

if you are using blockchain beta driver it will do better 22-24, or you can use drivers from 17.10.2 (you'll need to set "compute" mode for each card) 22-24MH and if you mod you'll get 28MH approx.

expedition is somehow better then Strix, yet its the slowest 570 I've got, here is a dozen on h110 btc+:

[jackbox]

hi guys,

I just built a new rig, Fatal1ty H270 Performance with 4GB ram, I3 cpu and 6 x asus radeon rx 570 expedition oc

When I installed 2 cards first on their own in PCIe slots, then windows 10 etc , blockchain drivers plsu patchers I would get solid 27MH/s (BIOS moded on all cards).

then, I moved the mobo to open case and plugged all six GPUs in using PCIe risers, installed drivers etc, then used the patcher, I can see in GPUz all cards have correct drivers etc.

I run Claymore , it detects 6 cards, stars mining and all I get is 21mh/s

any idea what am I doing wrong please?

if you are using blockchain beta driver it will do better 22-24, or you can use drivers from 17.10.2 (you'll need to set "compute" mode for each card) 22-24MH and if you mod you'll get 28MH approx.

expedition is somehow better then Strix, yet its the slowest 570 I've got, here is a dozen on h110 btc+:

Are the new drivers compatible with Afterburner?

[Victorio]

expedition is somehow better then Strix

Sorry for the offtopic, but what's your statistics on Asus GPUs? I have heard nasty opinions on them - some people report fault rates up to 7% on Asus models and I've met more complaints over forums vs other manufacturers.
Just in case, no intention to FUD against Asus, been using Asus MBs for over 10 years.

[joojoohe]

Hi guys.
i'm mining Ethereum on Nanoppol.
What mean this countdown (on my dash on nanopool)?? : Next DAG Epoch 1 : 12 : 48 : 34

[glester]

Next DAG change, difficulty adjustment.

[Rickirs]

I have tried mining with Claymore miner v10.2 but have being getting high stale shares of 10 to 12%.
When I was using v9.8, my stale shares are only about 4%.
Any advice?

Thanks

Not sure but I have hardly any stale share with v10, installed 10.2 today, so fare no stale there too.

[Rickirs]

Received this error today on v10.2:

Code:

GPU 0 mining was stopped because -tstop temperature reached.

Its ASUS RX 580 8G, however if I check temperature, its just 53C, even fan is not spinning as aroudn 55C, this its fan stops automatically.

Code:

====================    ROCm System Management Interface    ====================
=====================================================================
  0   67df   53.0c    58.14W   600Mhz   2000Mhz  0%   auto      0%       N/A
======================================================================
====================           End of ROCm SMI Log          ====================

But if I use Claymore v10, then it works .

[garybeng]

I have tried mining with Claymore miner v10.2 but have being getting high stale shares of 10 to 12%.
When I was using v9.8, my stale shares are only about 4%.
Any advice?

Thanks

Not sure but I have hardly any stale share with v10, installed 10.2 today, so fare no stale there too.

What is your rig setup?
I am using Asus Maximus Hero8 with i3-6100. Corsair LPX 2x8GB RAM, 128GB M.2. ColorPower 2xRX580, 2xRX580 MSI Gaming X, 2xAsus Strix R390 OC Gaming.

I also realized that I have lower than 3% stale shares if I run 4xGPU but higher than 5% stale shares on 6xGPU on v9.8
What can be the problem? How can I reduce stale shares?

[Cryptographertm]

why doesnt nofee work anymore

I was trying to test some setting and every time i opened the thing it started mining for the dev. like 5 times in a row

i added nofee restarted came back and its mining for the dev again..

[SackofFluor]

Hey Guys


I have a little problem...

After some tinkering i was able to bring my RX580 8GB cards to stable 29-30Mh/s. I have 5 of those cards and 1 of the 4GB version (which is also pretty good i have to admit).

But now to my problem, if i start the cards all individually they all bring the same performance, maybe +- 0,1 Mh/s, but when i start all of them together two of my cards need longer for the DAG creation 7200ms instead of the other 4900-5500ms .

And they only hash at around 20-22Mh/s.... I have no idea what i can do to bring some up with their buddies and like written further up.... on their own they are entirely capable of going 29-30 ....

Does anybody has any idea what i can do to improve my little problem ?



Sincerely

Sack

[Rickirs]

I have tried mining with Claymore miner v10.2 but have being getting high stale shares of 10 to 12%.
When I was using v9.8, my stale shares are only about 4%.
Any advice?

Thanks

Not sure but I have hardly any stale share with v10, installed 10.2 today, so fare no stale there too.

What is your rig setup?
I am using Asus Maximus Hero8 with i3-6100. Corsair LPX 2x8GB RAM, 128GB M.2. ColorPower 2xRX580, 2xRX580 MSI Gaming X, 2xAsus Strix R390 OC Gaming.

I also realized that I have lower than 3% stale shares if I run 4xGPU but higher than 5% stale shares on 6xGPU on v9.8
What can be the problem? How can I reduce stale shares?

Asus B250 Mining, Pentium G4400 , 1x8GB Hyperx Fury, Corsair RM1000 , 4x1060 (Asus, Zotac), 2 x Asus RX580 Dual OC

However, my Asus 580s are not yet modded or overlocked,they are hashing at 22Mhz atm but will be modding soon.

[Rickirs]

Hey Guys


I have a little problem...

After some tinkering i was able to bring my RX580 8GB cards to stable 29-30Mh/s. I have 5 of those cards and 1 of the 4GB version (which is also pretty good i have to admit).

But now to my problem, if i start the cards all individually they all bring the same performance, maybe +- 0,1 Mh/s, but when i start all of them together two of my cards need longer for the DAG creation 7200ms instead of the other 4900-5500ms .

And they only hash at around 20-22Mh/s.... I have no idea what i can do to bring some up with their buddies and like written further up.... on their own they are entirely capable of going 29-30 ....

Does anybody has any idea what i can do to improve my little problem ?



Sincerely

Sack

I have noticed same but only difference is 0.5 to 0.8 Mhz in my case if I run multiple cards, even have to lower clock a little to keep things stable, though I have 1060 and RX580s mix on same rig. But with 8-9 Mhz difference, that is odd. Have considered your PSU , is it not able to distribute power or its underpowered somehow for all running at same time ?

BTW, I don't think DAG creation time has any issue, with me, they usually takes around ~7000-8000 for each , yet all working fine.

[Rickirs]

Just reporting about AMDGPU-PRO v17.50 drivers.
http://support.amd.com/en-us/kb-articles/Pages/Radeon-Software-for-Linux-Release-Notes.aspx
http://support.amd.com/en-us/kb-articles/Pages/Installation-Instructions-for-amdgpu-Graphics-Stacks.aspx
The driver architecture seems to have evolved, and some option seems to install new opensource components (from Vulkan, ROCm OpenCL libs - Phoronix make article on that), but you can install legacy version.

I've installed with deprecated option on latest Ubuntu LTS update 16.04.3, kernel 4.10
./amdgpu-pro-install -y --compute
(in fact: ./amdgpu-pro-install -y --headless --opencl=legacy )
It finally worked nice, with no change in performance

however big warning: do clean uninstall before. I had initially installed the full amdgpu-pro with X11, then the --compute version headless, and when I installed v17.50 in compute mode, it could not compile the driver, and there was broken dependencies in apt-get... finally I hade to use dpkg --purge on all amdgpu related components, especially X11 libs, opencl libs...

Anyway, I see no improvement. is there some improvement to expect with v17.50 ? I use -mode 1 -asm 2 -etha 1 fro RX480/580... is there some prefered settings today ?

best regards.

I was on 17.40 , updated to v17.50 , installed without any problem however, my GPU (RX580) was not detected. I can see it lspci | grep VGA but claymore can't see , neither rocm-smi can see it.

Removed it and installed 17.40, now it can't install, giving some kernel related error that this is not made for this 4.11 (rocm kernel). So I don't have any AMDGPU-pro driver but on reboot card is detected and its working with same hash I was getting previously.

How following shows nothing.

Code:

dpkg -l amdgpu

But

Code:

lshdw -c video 

shows amdgpu driver loaded for my GPU , how do I find which version is that ?

[NewbieMiner17]

Hey guys,
me again!

So. I'm still running 6x RX480 4gb saph nitro+. Blockchain drivers, Memshift Anorak Bios, OC with AB.

So out of two rigs with exactly the same setup I've recently been mining on nicehash with around 28MH per card. On Nanopool dual mining ETH/SC I've been getting around 25MH. After the whole nicehash hack I've had to go back to nanopool until they get the systems back up and running.

Thats all fine and one rig is running golden. My other rig however has now decided that it will not mine at anything higher than 19MH with the same settings. I've tried reinstalling the drivers, reupdating the Bios, playing with OC, went back to V10 claymore, etc but no luck the cards are just refusing to run at over 19MH. Does anyone have any idea what could be causing this? I think I'm going to switch the hard drives around tonight to see if that does anything but short of that I'm stumped. Guess it would mean a fresh windows install if it does solve it.

PS. when I tried powersave bios on half the cards it did drop it even further to 16MH. previously on powersave I was getting around 22.

Looking forward to hearing what ideas you guys might have.

[Godmode45]

Have the same problem,  8xrx570+blockchain drivers. mining only at 20mh..
When I look at system ram, claymore use all 8gb.
Claymore restart after some time and the cycle starts over.
My other 6gpu rigs claymore use 450mb, anyone know why this is?

[BSAlex]

Hey guys,
me again!

So. I'm still running 6x RX480 4gb saph nitro+. Blockchain drivers, Memshift Anorak Bios, OC with AB.

So out of two rigs with exactly the same setup I've recently been mining on nicehash with around 28MH per card. On Nanopool dual mining ETH/SC I've been getting around 25MH. After the whole nicehash hack I've had to go back to nanopool until they get the systems back up and running.

Thats all fine and one rig is running golden. My other rig however has now decided that it will not mine at anything higher than 19MH with the same settings. I've tried reinstalling the drivers, reupdating the Bios, playing with OC, went back to V10 claymore, etc but no luck the cards are just refusing to run at over 19MH. Does anyone have any idea what could be causing this? I think I'm going to switch the hard drives around tonight to see if that does anything but short of that I'm stumped. Guess it would mean a fresh windows install if it does solve it.

PS. when I tried powersave bios on half the cards it did drop it even further to 16MH. previously on powersave I was getting around 22.

Looking forward to hearing what ideas you guys might have.

Please check what EPOCH is used in both cases. I have a miner that use different EPOCH randomly, and mining speed is completely different: from 9.5Mh to 23Mh (!!!). Did not find any possibility to control what EPOCH is used so in my case the problem is unresolved :-(

[AlainC]

I was on 17.40 , updated to v17.50 , installed without any problem however, my GPU (RX580) was not detected. I can see it lspci | grep VGA but claymore can't see , neither rocm-smi can see it.

Removed it and installed 17.40, now it can't install, giving some kernel related error that this is not made for this 4.11 (rocm kernel). So I don't have any AMDGPU-pro driver but on reboot card is detected and its working with same hash I was getting previously.

How following shows nothing.

Code:

dpkg -l amdgpu

But

Code:

lshdw -c video 

shows amdgpu driver loaded for my GPU , how do I find which version is that ?

to know the package relative to amdgpu try
" dpkg -l| grep amdgpu|grep -v ubuntu16"
(the grep -v ubuntu16 is because some adgpu libs are from core ubuntu, not amdgpu-pro. check each name)
but there are also many other pakage related but without the name
then do somethin like
"dpkg --purge amgdpu-core amdgpu-pro-core  amdgpu-dkms ... and so on"
or if you are lazy and feel lucky (check the list generated before)
"dpkg --purge `dpkg -l| grep amdgpu|grep -v ubuntu| awk '{ print $2;}'` "
check if there are pachage with failed dependencies with
"apt-get install"
if there are, purge them too...

and then reinstall (--opencl=legacy and probably --headless unless you use X11 on your GPU).

I have a Sapphire RX580, so it should work...
 

[cryptotore]

Has anyone been able to get claymores miner to work on antpool?

I know that "Pools that require "Login.Worker" instead of wallet address are not supported directly currently, but you can use "-allpools 1" option to mine there."

But I'm not really sure how this is going to help me?