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

[Balthazar]

- fixed issue with "-esm 3" (again).

Sorry for disturbing you again, but it looks like -esm 3 mode is still not stable.
People are reporting sporadic app crashes while using -esm 3.
Same systems are perfectly stable while using -esm 0 and can work during the weeks without any issues.
Memory corruption error, maybe?

I'm not sure if it is related to "-esm 3". Also I cannot fix it with this screenshot, I need some related log files and some way to reproduce it.

Hi.

One of users uploaded this log file:

https://drive.google.com/file/d/1EhuT77U3SiCo3K4e_DY9MvFSpijoK2cD/view?usp=sharing

[1714843269]

1714843269

[1714843269]

1714843269

[1714843269]

1714843269

[1714843269]

1714843269

[1714843269]

1714843269

[1714843269]

1714843269

[MaxMidnite]

Claymore gone very quiet these days, hope everything is ok?

[Claymore]

One of users uploaded this log file:
...

Hmm, I don't see any problems in this log file. If it's a bug, it's difficult to catch it since I cannot reproduce it, the log is ok and it takes a lot of time to get it (about 80 hours in this log). It's not related to the running time directly since I have some systems running >200 hours completely stable. Another thing is that you can get this error if drivers fail due to some reason, in most cases OpenCL call just hangs so watchdog can catch it, but sometimes it can crash entire application or entire system.
I will try to find the reason, but the best solution currently is to restart miner from .bat file automatically (or better reboot the system). As far as I remember Windows has a way to disable that dialog about crash...
I will also put some rigs to your pool in "-esm 3" mode to check if I can see this error.
BTW, tell to your user that the miner detected devfee cheating and works a bit slower as described in Readme. Some people ignore Readme or think that I'm not smart enough to detect public tools, private networks, local proxies, etc. Be smarter if it's really important for you, or use Linux version, it's free of any cheat detection.

Claymore gone very quiet these days, hope everything is ok?

I'm fine, thanks, is v11.7 too old and you want v11.8 right now?

[mettalmag]

is there any chance to add to monitoring software the ICMP/ping function?
when remote restarting (in my case relay switches) the rig instead of RDPing in it or using VPN to monitor the rig is responding or not one just can simply use remote from HTTP and see if the rig appeared online or not after restarting, I hope this is not only my concern
cheers

[Claymore]

is there any chance to add to monitoring software the ICMP/ping function?
when remote restarting (in my case relay switches) the rig instead of RDPing in it or using VPN to monitor the rig is responding or not one just can simply use remote from HTTP and see if the rig appeared online or not after restarting, I hope this is not only my concern
cheers

You can connect to miner via http or json to check if it's alive ("-mport" option).
I think also about an option to start some .bat file from miner every X seconds, but not sure if it will be really useful.

[Balthazar]

One of users uploaded this log file:
...

Hmm, I don't see any problems in this log file. If it's a bug, it's difficult to catch it since I cannot reproduce it, the log is ok and it takes a lot of time to get it (about 80 hours in this log). It's not related to the running time directly since I have some systems running >200 hours completely stable. Another thing is that you can get this error if drivers fail due to some reason, in most cases OpenCL call just hangs so watchdog can catch it, but sometimes it can crash entire application or entire system.
I will try to find the reason, but the best solution currently is to restart miner from .bat file automatically (or better reboot the system). As far as I remember Windows has a way to disable that dialog about crash...
I will also put some rigs to your pool in "-esm 3" mode to check if I can see this error.
BTW, tell to your user that the miner detected devfee cheating and works a bit slower as described in Readme. Some people ignore Readme or think that I'm not smart enough to detect public tools, private networks, local proxies, etc. Be smarter if it's really important for you, or use Linux version, it's free of any cheat detection.

Another report with log file:

https://forum.bits.media/index.php?/topic/33998-eth-%D0%BF%D1%80%D0%BE%D1%81%D1%82%D0%BE%D0%B9-ethereum-%D0%BF%D1%83%D0%BB-pps-pot/&page=363&tab=comments#comment-1435514

[Claymore]

Another report with log file:
...

More interesting because it took only an hour to crash, but still no details in the log file. I will test some rigs in "-esm 3" mode for 48 hours to check if I can reproduce this issue here. If I can I will fix it immediately, otherwise it can take a lot of time to find the reason.

[DireWolfM14]

I just switched from Phoenix to see how dual mining will work for me.  I'm fairly new to all this, I'm just starting out with a pair of MSI GTX GeForce 1070s (armor.)  My ETH hash rate is 30.06MH/S per gpu.  That's 2.1MH/S slower than it is with the Phoenix miner.  I didn't turn the dev fee off.  Is this normal?

Does dual mining make up for the difference in hash rate and dev fee over Phoenix?  I'm mining ETH and DCR, and seeing results on my pools' dashboards, so it's working.  I guess I'll run it for a couple of days and find out.

[xunl]

I also get these errors repeatedly on 11.7. Each time claymore miner will crash & restart when error occurs. Anything I can do to avoid them?
The these errors are for gtx 1070/1060 6GB

Code:

23:42:31:285	668	em hbt: 15, fm hbt: 31, 
23:42:31:285	668	watchdog - thread 0 (gpu0), hb time 234
23:42:31:285	668	watchdog - thread 1 (gpu0), hb time 109
23:42:32:769	8e4	GPU 0, GpuMiner cu_k1 failed 4, unspecified launch failure
23:42:32:769	8e4	GPU 0, GpuMiner kx failed 1
23:42:32:769	8e4	Set global fail flag, failed GPU0
23:42:32:769	8e4	GPU 0 failed
23:42:32:769	8e0	GPU 0, GpuMiner cu_k1 failed 4, unspecified launch failure
23:42:32:769	8e0	GPU 0, GpuMiner kx failed 1
23:42:32:769	8e0	Set global fail flag, failed GPU0
23:42:32:769	8e0	GPU 0 failed

Code:

GPU 0, GpuMiner cu_k1 failed 77, an illegal memory access was encountered
GPU 0, GpuMiner kx failed 1
Set global fail flag, failed GPU0
GPU 0 failed
GPU 0, GpuMiner cu_k1 failed 77, an illegal memory access was encountered
GPU 0, GpuMiner kx failed 1
Set global fail flag, failed GPU0
GPU 0 failed

Code:

02:29:01:610	1390	GPU 0, GpuMiner cu_k1 failed 73, an illegal instruction was encountered
02:29:01:610	1390	GPU 0, GpuMiner kx failed 1
02:29:01:610	1390	Set global fail flag, failed GPU0
02:29:01:610	1390	GPU 0 failed
02:29:01:610	1160	GPU 0, GpuMiner cu_k1 failed 73, an illegal instruction was encountered
02:29:01:610	1160	GPU 0, GpuMiner kx failed 1
02:29:01:610	1160	Set global fail flag, failed GPU0
02:29:01:610	1160	GPU 0 failed

[nitrobg]

Is it normal to have a slightly lower hash on AMD cards with Linux(HiveOS) compared to Windows 10 on identical settings (-asm 2 -ethi 16) and overclock?

Windows:

Code:

GPU0 27.459 Mh/s, GPU1 29.877 Mh/s, GPU2 30.349 Mh/s, GPU3 28.585 Mh/s

Linux:

Code:

GPU0 27.101 Mh/s, GPU1 29.517 Mh/s, GPU2 29.930 Mh/s, GPU3 28.406 Mh/s

[skaraud]

Hi, I'm using Claymore on windows since a year, and recently I have a loss of my reported hashrate on NanoPool (ETH Pool) with the latest 11.7 version. Is anyone having the same issue ? I mean I have a correct reported hashrate during 5 or 6 hours, then it drops to zero but the rig continues mining at a stable hashrate. it's just that I have no reported hashrate on my pool monitoring interface.

[Stoukie]

Hi, I'm using Claymore on windows since a year, and recently I have a loss of my reported hashrate on NanoPool (ETH Pool) with the latest 11.7 version. Is anyone having the same issue ? I mean I have a correct reported hashrate during 5 or 6 hours, then it drops to zero but the rig continues mining at a stable hashrate. it's just that I have no reported hashrate on my pool monitoring interface.

Have exactly the same issue on Nanopool, not a Claymore issue but I suspect issues on the pool itself... Switched to Ethermine.org for now and everything is working 100%.

[Stoukie]

Good day guys,

Made an interesting observation yesterday after playing around with my .bat file, running 5 x RX580 8GB's and getting around 29 to 30 MH/s per card.

Example of the .bat file I have been using:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

EthDcrMiner64.exe -epool <MyPoolAddress> -ewal <My ETH Address / My Email Address> -epsw x -allpools 1 -allcoins 1 -gser 3 -lidag 1 -cclock 1150 -cvddc 950 -mclock 2115 -mvddc 950 -ethi 15 -etht 20

Then I switched over to a vanila .bat file only specifying the minimal and my hashrate actually went up to around 30.5 MH/s:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100

EthDcrMiner64.exe -epool <MyPoolAddress> -ewal <My ETH Address / My Email Address> -epsw x

I have not checked yet but I presume the undervolting made a difference in electricity usage per card, but 'overclocking' did nothing for my hashrate.

Just an observation from my side...

[ermis1975]

If you mod bios, why you have to play with start bat file settings, of wattman, afterburner etc ? Or you dont mod bios and just change settings manually?

[leonix007]

If you mod bios, why you have to play with start bat file settings, of wattman, afterburner etc ? Or you dont mod bios and just change settings manually?

Some modding bios are just simply copy and pasting straps from lower freq, and didn't play along with the settings

there are pros and cons of it, depending on your usage

some may prefer doing adjustments thru overclocking tools such as wattman, OD, MSI AB, as they could play it on different algos

[Lunga Chung]

is there any chance to add to monitoring software the ICMP/ping function?
when remote restarting (in my case relay switches) the rig instead of RDPing in it or using VPN to monitor the rig is responding or not one just can simply use remote from HTTP and see if the rig appeared online or not after restarting, I hope this is not only my concern
cheers

which relay switch are you using?

[Andre438]

hello, tell me on which coin lutshe set up mined dcr but came out visibly ASIC on this algorithm and production stopped:( mined at decredpool.org the withdrawal from 1 coin I managed to get 0.38 I am writing through a translator. 2 gpu  rx580 Rx460

[jamfro]

Was wondering if Claymore had a chance to look at this problem of not being able to use more than 6 amd gpus after 1803 upgrade.  And hoping he can provide a projection of if and when he can work around whatever was borked by the "upgrade".

Thanks

[Geniusvzla]

Hi, I'm using Claymore on windows since a year, and recently I have a loss of my reported hashrate on NanoPool (ETH Pool) with the latest 11.7 version. Is anyone having the same issue ? I mean I have a correct reported hashrate during 5 or 6 hours, then it drops to zero but the rig continues mining at a stable hashrate. it's just that I have no reported hashrate on my pool monitoring interface.

hello, in fact it is also reporting 0 in my case, but it seems to be an inconvenience in the pool, let's hope to see if they fix it

[mine_phun]

Hello All,

As admin of NLpool i wonder how you guy's think about Dualmining with claymore after 9th May.
As you probably all are aware XVG will halve around that date, @ block 2.124.000 to be precise.
This means that when you dual mine XVG, your income for XVG will halve also.

From that date i think, with the current exchange rate, XVG dual mining will not be profitable anymore unless the price for XVG will double or half of the miners go out, something i personally don't see happening soon.

So my question is what are you guy's going to do? and what do you want us to do?

For now we have Shield as alternative for blake2S and Maxcoin / Smartcash with the Keccak algo.

However those coins cannot take the 250TH/s from Verge so if you guys have alternatives you would like to see, i would really like to know them if you can share. Perhaps someone can make a poll around here?