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

[ghost2132]

I am having an issue after I extract the Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v11.1 for Windows.  I use 7-zip and after I extract the files I cannot run or access any of the files.  I did notice that 7-zip says the host os uses fat while I have NTFS.  Should I use a different program to extract the files?  I did try winrar as well.

Thank you.

Have you downloaded the program from the oficial link?

https://mega.nz/#F!O4YA2JgD!n2b4iSHQDruEsYUvTQP5_w

Yes I did.  Other zips I download work fine, it is just the newer dual miner that I have this issue.

In case it helps anyone else, with 7-zip I had to do a copy to instead of the normal extract files for the files to work.

[1714112473]

1714112473

[1714112473]

1714112473

[nitrobg]

There are no AMD miners for X17, this algo gets quite popular lately. Perhaps Claymore could fill that hole?  

[The Godfather]

is it safe to use 10.2?
I have to much stale shares with new claymore..
Maybe its not up to miner, but stale shares went hight.. from 10% to 90%

[Claymore]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

[ErikETHMiner87]

Hello everyone,

I hope I can find a solution to this problem here, it seems that I am not the only one that has gone through this. (I have yet to fix with solutions I've seen so far)

First I will show you what I am running so far:

Windows 10 Home 64 Bit OS
1 Gigabyte GTX 1070 WINDFORCE OC 8GB  (4 others on the way, so I'm freaking out) - It is on a powered riser.
1200 W EVGA Titanium 80 Plus PSU Fully Modular
MSI Z270 A PRO MOBO
Celeron Processor
4GB DDR4 RAM
64GB SSD

My rig starts normally and begins to mine with no issues ,  then either 40 minutes later or 2 hours later the computer crashes and shuts down completely.

Error Message in red text:
NVML: cannot get current temperature, error 999 (an internal driver error occurred)
WATCHDOG: GPU 0 Hangs in OpenCL call, you need to restart miner  Sad
WATCHDOG: GPU 0 Hangs in OpenCL call, you need to restart miner  Sad
ETH: No new jobs were received for 10 mins - something is wrong with pool

Things I have tried:
1. Switched powered risers on the card
2. Re-installing NVIDIA drivers to the latest version including CUDA 9.1
3.Updated MSI BIOS
4. Kept OC settings to default.
5. Changed DNS servers to Googles
6.Manually connected to Stratum using Nanopools IP.

Should I dump nanopool and claymore entirely and move on to another software ?

Please if anyone out there has fixed this issue any help would be much appreciated I am completely stuck.

[et0nnic]

Can confirm I see an 1Mh/s overall increase on my AMD rig with v11.2.

[AnViGo]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

I have rx480+4x470x, the perfomance drop 1-1,2MH/s per card in v11.2

[joaocha]

@Claymore,

I found a bug on dual mining XVG,

both on Rx 570 and 580

if you disable xvg mining on the fly(pressing the numbers), the Eth speed drops a little(1 or 2%), and only comes back to normal if you reenable the dual mining

[Claymore]

@Claymore,

I found a bug on dual mining XVG,

both on Rx 570 and 580

if you disable xvg mining on the fly(pressing the numbers), the Eth speed drops a little(1 or 2%), and only comes back to normal if you reenable the dual mining

Use "+" and "-" keys to find best -dcri value for ETH-only mode. The value can be different from best -dcri value for dual mining mode.

[Vann]

Nice. With 11.2 I got a 3 MH/s bump on ETH dual mining on a 13 card RX 480/580 rig and 1 MH/s on a 6 card RX 570 rig. Thanks!

[cero10]

Guys is there a way to mod a RX580 so it can mine eth ath 2000+ mclock? everytime i go more than 2000 memory speed windows crashes. Is there a way to add another memory strap at 2250?

[Vann]

It depends on the memory type. Even with good timings Samsung is hard to overclock over 2050 MHz stable and Hynix/Micron can easily overclock to 2200-2250 MHz depending on the memory straps and ASIC quality. The best you can do is find/buy optimized timings for your memory type.

[The Godfather]

from 30.2 to 30.5 but we will se if there is less stale shares..

It depends on the memory type. Even with good timings Samsung is hard to overclock over 2050 MHz stable and Hynix/Micron can easily overclock to 2200-2250 MHz depending on the memory straps and ASIC quality. The best you can do is find/buy optimized timings for your memory type.

but samsung can be more undervolt. you can lover voltage more on samsung then other 3

[LucianoBM]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

Thank you for the updates!!

This time I had a decrease on 2 MH/s on each RX 570 (I have 6 on this rig).
Any idea why?

[AnViGo]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

Thank you for the updates!!

This time I had a decrease on 2 MH/s on each RX 570 (I have 6 on this rig).
Any idea why?

I have same problem, try to increase dcri up to 40
It works in ETH-only mode

[Claymore]

You perhaps need to readjust -dcri for ETH-only mode, use "+" and "-" keys for that and "s" key to see current speed. You can also try "-asm 2" option with various -dcri to see if it is faster.

[Bael]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

the hashrate has gone up !! perfect : )

[Vann]

from 30.2 to 30.5 but we will se if there is less stale shares..

It depends on the memory type. Even with good timings Samsung is hard to overclock over 2050 MHz stable and Hynix/Micron can easily overclock to 2200-2250 MHz depending on the memory straps and ASIC quality. The best you can do is find/buy optimized timings for your memory type.

but samsung can be more undervolt. you can lover voltage more on samsung then other 3

Undervolting the memory has very little effect in power consumption compared to underclocking and undervolting the core. Undervolting also depends more on ASIC quality than anything else.

[groovytechno]

v11.2:

- some improvements in AMD GPU kernels, it can increase hashrate a bit in some cases.
- reduced some delays, it will cause a bit less stale shares.

Excellent update, Claymore.  Each of my rigs, depending on the cards have collectively reported roughly 1-3 Mh/s more.    Thank you!!

[naldo23]

Hello guys! I have one problem I set my 580 8GB rigs to do dual mining on black2s with an dcri 45 and I don't have any hashrate drops from ethereum but when I try to set the cards 580/570 with only 4GB I can make a proper settings to don't have drops on my ethereum hashrate evan if I set dcri to 20.. How we can set the 4GB cards to don't have drops? any idea guys?

With dcri 20 I still have less with around 10-5 MH/s on 6 570 4 GB cards. Thanks!

Nobody have this issue guys? Maybe you Claymore can help me with this...