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

[bitdag]

I used 2 days 11.2 +1mh and %4 %5 stale it is ok for me but today 25% stale switched to previous ver 11.0

[1713548476]

1713548476

[1713548476]

1713548476

[Ursul0]

11.2 must have some hardcoded OC settings. Suddenly I get GPU errors, more heat, more instability. On some GPU it works but some not.

I want to do my own OC settings! Remove OC and release 11.3

I couldn't agree more! As it is, I'm planning on using 11.1 for a good long time. At least it's stable...

this is some weird claim. I have no idea what it means nor probably the poster.
v 11.2 looks a bit unstable at first, but now it seems fine and I'm getting almost 1.5% hash increase across my polarises

[SnOOpeRa]

- EthMan loses rigs with 12 GPUs.
  Sometimes systems with 12 GPUs and low-end CPU become slow for remote access, you can see problems with EthMan and other remote management software.

I know there are already posts about this problem , but did not found workaround , on mainboard with 12 polaris gpus win 10 1709, and claymore dual miner ( tested with 11.2 down to 10.6 ) ethman do not work when all 12 gpus are mining , lost connection immediately , when start just 8 gpus to mine ethman works without problem , it's not cpu problem , cpu load stays 20-30% , remote software is working good ( teamviewer , splashtop , tightvnc ) , i tryed phoenix miner 2.7c recently which works with ethman and with it there is no problem when 12gpus are mining , cpu stays the same load 20-30% , so the problem must be with claymore's miner socket implementation , web server or something like that when gpus are more than 8 , or some hardcoded setting remain from the times when driver and windows accept only 8 gpus . Can you check and fix this problem please .

p.s: I read through the posts and did not find workaround for this , so if there is one , my apologies.

[GoBlack]

I have the same problem, 0 memory errors but I always have a 6/7% stale share on 11.2, ethermine pool :S

Yep same here. Got 4-7% stale shares.

I have stale shares about 5% to 10% on 11.0 also, moved to 11.2 still the same amount of stale shared on Ethermine.org... Move to Nanopool then they will not bother you... as they do not report on stale shares 

11.2 was a failure but 11.1 is stable (at least for me). As far as stale shares go, you can reduce them by checking for work faster than what is set by default. Add the switch -etht which by default is set to 200ms and reduce the time your rigs wait to check for shares. I have mine set to check every 20ms with this switch -etht 20

In my experience this keeps stale shares under 5% (Mine is currently at 2% using 11.1). But you can do some pings to make sure that your problem isn't latency between your router and the pool or even latency on your own network as well.

What is your ping to the pool?

mine is 95ms, do you think I see any difference?

My ping is around 40ms right now but yours seems fast enough. Give -etht 20 a try for a day and see if the stale shares go down for you.

[whoismoses]

I switched 6 of my 8 rigs to 11.2 from 11.0 (I've been busy).

So far stale shares look just fine. It's been about 24 hours. All AMD Polaris cards.

[EvilHamster]

Firstly, thank you all for your help so far (messages and so on)

My rig - now 4 weeks old, ALL new kit.

Tried 11.2 today and apart from virus guards kicking off all over the place (until I disabled them all) failed to work.

I have now kind of got the rig working thanks to some help, however...

Starts at 64 Mh/s then reduces down and down...

ETH: GPU0 21.917 Mh/s, GPU1 10.142 Mh/s, GPU2 10.124 Mh/s, GPU3 21.923 Mh/s


ETH: GPU0 0.000 Mh/s, GPU1 10.132 Mh/s, GPU2 10.176 Mh/s, GPU3 21.906 Mh/s



Followed shortly by:

WATCHDOG: GPU 0 hangs in OpenCL call, exit
WATCHDOG: GPU 0 hangs in OpenCL call, exit
Quit, please wait...
Restarting OK, exit...

Then rig comes to a complete stand still...Wont go further.

Virtual memory 16gb, virus guard disabled...

Please any thoughts - this is killing me now...

Rig

Celeron - K170 board - 4gb ram - Msata HDD - 2 x rx570 4gb - 2 x RX550 4gb - Thermaltake 850psu...Windows 10 64bit.

yup... i posted the exact same issue a few days back but never saw a reply with fix suggested. Here's the thing... this is happening on all 4 of my 6x470 rigs... and really only since that last major W10 update.  So, I am saying it is NOT a hardware issue because what are the odds that on 4 rigs, there is an actual problem with GPU0 on all 4?  It never happens on any other GPU, so this means it is a software/settings issue, but I have not been able to figure it out.  On Claymore Cryptonite miner, the solution to fix a lagging GPU0 was to manually designate the "-h" value in the .bat file, but no such option existing in this miner, at least not -h but maybe something else can be tweaked only on GPU0 to fix this... that has been what i've tried to discover for the past weeks but no solution yet. I actually had to disable GPU0 on 1 of my rigs that it got so bad, and looks like another 1 is headed that direction as well.

The one difference with mine is that they don't actually crash/freeze, but they get stuck in a non-stop look of resarting the miner due to that hanging GPU0 issue. So it restarts every couple minutes, usually with GPU0 at zero speed.

Claymore... or anyone else that knows of this issue... please advise. Thanks!

Windows 10 Pro - 8GB RAM - 16-32GB virtual memory - Claymore dual miner versions 10.2/10.5/10.6

Do you have a dummy plug or a monitor connected to GPU0 ? I have noticed this behavior when you remote and don't have a monitor/dummy plug connected

Nope - I have now stripped rig back to 2 x RX570;s nothing plugged in - no windows update - it ran perfect on 10.6 - jumped 11 and 11.1 now tried 11.2 - blat..

Tried re-installing drivers and all sorts.

GPU0 was doing 0 MHS so I now have taken GPU 1 which was running around 21-22.

ETH: GPU0 0.000 Mh/s
GPU0 t=54C fan=23%%
ETH: 03/06/18-20:21:44 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:01
ETH: GPU0 0.000 Mh/s
ETH: 03/06/18-20:21:46 - New job from eu1.ethermine.org:4444
WATCHDOG: GPU 0 hangs in OpenCL call, exit
GPU0 t=54C fan=0%%
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:02
ETH: GPU0 0.000 Mh/s
Quit, please wait...

So now feck all is happening in GPU0 regardless of whichever card you try.

[GoBlack]

Firstly, thank you all for your help so far (messages and so on)

My rig - now 4 weeks old, ALL new kit.

Tried 11.2 today and apart from virus guards kicking off all over the place (until I disabled them all) failed to work.

I have now kind of got the rig working thanks to some help, however...

Starts at 64 Mh/s then reduces down and down...

ETH: GPU0 21.917 Mh/s, GPU1 10.142 Mh/s, GPU2 10.124 Mh/s, GPU3 21.923 Mh/s


ETH: GPU0 0.000 Mh/s, GPU1 10.132 Mh/s, GPU2 10.176 Mh/s, GPU3 21.906 Mh/s



Followed shortly by:

WATCHDOG: GPU 0 hangs in OpenCL call, exit
WATCHDOG: GPU 0 hangs in OpenCL call, exit
Quit, please wait...
Restarting OK, exit...

Then rig comes to a complete stand still...Wont go further.

Virtual memory 16gb, virus guard disabled...

Please any thoughts - this is killing me now...

Rig

Celeron - K170 board - 4gb ram - Msata HDD - 2 x rx570 4gb - 2 x RX550 4gb - Thermaltake 850psu...Windows 10 64bit.

yup... i posted the exact same issue a few days back but never saw a reply with fix suggested. Here's the thing... this is happening on all 4 of my 6x470 rigs... and really only since that last major W10 update.  So, I am saying it is NOT a hardware issue because what are the odds that on 4 rigs, there is an actual problem with GPU0 on all 4?  It never happens on any other GPU, so this means it is a software/settings issue, but I have not been able to figure it out.  On Claymore Cryptonite miner, the solution to fix a lagging GPU0 was to manually designate the "-h" value in the .bat file, but no such option existing in this miner, at least not -h but maybe something else can be tweaked only on GPU0 to fix this... that has been what i've tried to discover for the past weeks but no solution yet. I actually had to disable GPU0 on 1 of my rigs that it got so bad, and looks like another 1 is headed that direction as well.

The one difference with mine is that they don't actually crash/freeze, but they get stuck in a non-stop look of resarting the miner due to that hanging GPU0 issue. So it restarts every couple minutes, usually with GPU0 at zero speed.

Claymore... or anyone else that knows of this issue... please advise. Thanks!

Windows 10 Pro - 8GB RAM - 16-32GB virtual memory - Claymore dual miner versions 10.2/10.5/10.6

Do you have a dummy plug or a monitor connected to GPU0 ? I have noticed this behavior when you remote and don't have a monitor/dummy plug connected

Nope - I have now stripped rig back to 2 x RX570;s nothing plugged in - no windows update - it ran perfect on 10.6 - jumped 11 and 11.1 now tried 11.2 - blat..

Tried re-installing drivers and all sorts.

GPU0 was doing 0 MHS so I now have taken GPU 1 which was running around 21-22.

ETH: GPU0 0.000 Mh/s
GPU0 t=54C fan=23%%
ETH: 03/06/18-20:21:44 - New job from eu1.ethermine.org:4444
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:01
ETH: GPU0 0.000 Mh/s
ETH: 03/06/18-20:21:46 - New job from eu1.ethermine.org:4444
WATCHDOG: GPU 0 hangs in OpenCL call, exit
GPU0 t=54C fan=0%%
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:02
ETH: GPU0 0.000 Mh/s
Quit, please wait...

So now feck all is happening in GPU0 regardless of whichever card you try.

Try it with Claymore 11.1 and see if you get the same results. My rig's did something similar because my bios settings min-max everything and the changes in 11.2 are just enough to push it off the cliff.
There are two ways to fix it 1) stop overclocking/undervolting manually and use the default bios and it works but with sub-par results compared to my own OC/UV settings or option 2) ...use Claymore 11.1

[_voyager_]

What -ethi value did you set for ETH solo miming mode?

[Begetan]

Claymor 11.1 and probably 11.2 constantelly crashes on AMD Radeon cards - RX580, 560, 460 in dual mode Etherium + XVG.
Linux - Ubuntu 16.04, latest standart version of Linux drivers. It's copletely different from SIA/LBRY mining.

Surprisely the same configuration are working good on Nvidia GTX1060-6G.

On Nvidia 1070/1070ti cards  enabling dual mining drops hashrate from 30-32 to 26-29 unpredictible way. It is not usable.  Miner should be tested and improved on this cards too.

Actually I missed much more income totally due to rigs hangs on than earn on dual mining of Verge.

[rigomortis]

I think Mr. Claymore is adding extra code to bypass all those nodev hacks.  I bet you each version he releases steals more shares from all you fucking lemmings.  

Mr. Claymore you have earned enough! Please release free version with no fees.  

[tenmoi]

Claymor 11.1 and probably 11.2 constantelly crashes on AMD Radeon cards - RX580, 560, 460 in dual mode Etherium + XVG.
Linux - Ubuntu 16.04, latest standart version of Linux drivers. It's copletely different from SIA/LBRY mining.

Surprisely the same configuration are working good on Nvidia GTX1060-6G.

On Nvidia 1070/1070ti cards  enabling dual mining drops hashrate from 30-32 to 26-29 unpredictible way. It is not usable.  Miner should be tested and improved on this cards too.

Actually I missed much more income totally due to rigs hangs on than earn on dual mining of Verge.

Running AMD too.
Cannot say anything about 11.1, but 11.2 is unstable. It crashed after 6 hours even though I had lowered mem and core clocks. If it crashes again, i'll go back to 11.0.

[cryptoillusion1]

Caaan u tell me without reading whole 1000 pages whats the diference betwen 9.6 vs current version? :> Is OpenCL.dll missing error thing is caused by old version?

You don't need to read 1000 pages - just download the latest software from the links on Page 1 - then read the History.txt file  which gives the revision history

[cryptoillusion1]

I think Mr. Claymore is adding extra code to bypass all those nodev hacks.  I bet you each version he releases steals more shares from all you fucking lemmings.  

Mr. Claymore you have earned enough! Please release free version with no fees.  

Interesting asking for a free version- should we then ask Microsoft,Apple,Samsung and others also to provide free ? would you work for free ? 

[roDIESEL]

I am puzzled.

There is a huge difference (shares) in what claymore reports and what pool reports.

Claymore Report / 1 hour:

https://imgur.com/dAcD4Ji

RiG01 > 193 Shares
RiG02 > 165 Shares

Ethermine Report / 1 hour:

https://imgur.com/1QUs2n5

RiG01 > 148 Shares (6 Stale)
RiG02 > 143 Shares (1 Stale)

...

Poolfee is 1%, claymore devfee is turned off.
This is quite a difference. Why is it like that?

Thanks for your feedback guys!

Would you bother to check the readme file, ESPECIALLY that part where Claymore says about what is happening when you disable hes fee ?

[rigomortis]

I think Mr. Claymore is adding extra code to bypass all those nodev hacks.  I bet you each version he releases steals more shares from all you fucking lemmings.  

Mr. Claymore you have earned enough! Please release free version with no fees.  

Interesting asking for a free version- should we then ask Microsoft,Apple,Samsung and others also to provide free ? would you work for free ? 

First off, Claymore uses open source code which he is trying to make a buck off.  So before you start kissing his ass.. or being his shill, get the facts right. 

Apple made OSX free upgrade.  Microsoft also made windows free.  Your point?

[roDIESEL]

My 570 8gb nitro got this kind of error..

OpenCL error -4 - cannot write buffer for DAG

even I change my Vram to 17500-25500

Is anyone of you have the solution for this? I already change also the riser, cable etc but I always got this message continuously..

I fixed an issue like that with a combo of more VRAM (64 GB from 32) and more RAM (8 GB from 4).
I am not sure though which one was the key, or if both were necessary .... point is that it works (for me).

[roDIESEL]

I think Mr. Claymore is adding extra code to bypass all those nodev hacks.  I bet you each version he releases steals more shares from all you fucking lemmings.  

Mr. Claymore you have earned enough! Please release free version with no fees.  

Interesting asking for a free version- should we then ask Microsoft,Apple,Samsung and others also to provide free ? would you work for free ?  

First off, Claymore uses open source code which he is trying to make a buck off.  So before you start kissing his ass.. or being his shill, get the facts right.  

Apple made OSX free upgrade.  Microsoft also made windows free.  Your point?

Point is that people are always asking for free good stuff, which is embarrassing.
Open source thing is also embarrassing as a reason. You want to make money, but you dont want to pay for your tool. Linux started as free open source operating system AND GUESS WHAT ! Some smart guys worked hard to get some special distributions (mission critical) that now COSTS MONEY !
Want me to make you a list with free miners that you may use instead of Claymore ?
Nobody is kissing hes ass, it is just common sense to pay a small fee for your making money tool, but somehow, you and other like you PRETEND to have hes work for free.
That is more than embarrassing. Its plain stupid.

[auscrash]

Just want to put my 2c in, I am running 4 separate rigs and have moved them all across to 11.2. I have seen the following:

• A noticable increase in reported hashrate on all 4 rigs (around 1-2mh/s on a 6 card rig)
• Although hard to specifically quantify, all 4 rigs seem to me to have a more stable hashrate, less dips.
• No reliability issues so far, all been running 24x7 without problems
• An increase in (nanopool) reported average hashrate on all rigs - I believe this comes from a combination of what i think is a more stable hashrate as well as the small increase seen with 11.2
• I needed to increase the DCRI setting on my rigs after moving to 11.2 for best results in my setups

I'm running RX580's on my rigs both 4GB and 8GB versions - so for me 11.2 is a big thumbs up, thanks Claymore 

[user1988]

I tried the drivers 18.3.1 and I can not control the fans

I want to use the fixed fans and color the -tt -50 and it does not apply

any solution?

[Vann]

Haven't tired th Adreniline 18.3.1 drivers but the 18.2.2 drivers are working with 11.2 for setting the overclock/undervolt/fan settings in Claymore and are very stable.

https://support.amd.com/en-us/kb-articles/Pages/Radeon-Software-Adrenalin-Edition-18.2.2-Release-Notes.aspx