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

[TR8888]

@ Claymore, Do you have or are you making a zcash gpu miner?

[Claymore]

Hey Claymore,

I am having some issues with some of the switches under Ubuntu
That would be a result of:
/home/ethos/claymore/ethdcrminer64 -lidag -di 012345 -eres 3 -tt 100 -epool us2.ethermine.org:4444 (...) etc.
Unless I am doing something wrong here - let me know one way or another.
Thanks,

"-lidag 1".

@ Claymore, Do you have or are you making a zcash gpu miner?

Search for "zcash" in this thread and you'll find my answer.

[armyman]

Hi everyone, I am mining eth solo with claymore 7.1 and I have OC my RX480 8 gb yesterday.

Everything about the software looks fine, but I was wondering:

1) can I exit the test mode on win10 without causing any crash?

2) what are the best settings in wattman to make sure that the cards have the lowest consumption possible and that they run as much stable as possible around the 31mh/s? Core, memory, fan, and temp? (at the moment they are varying between <24 and >30, which seems quite a lot)

this is the procedure I have followed to enable the test mode:

i) Press Windows Key + X to open Power User Menu. Select Command Prompt (Admin) from the menu.
ii) Once the Command Prompt open, enter bcdedit.exe -set loadoptions DISABLE_INTEGRITY_CHECKS and press Enter. Then enter bcdedit.exe -set TESTSIGNING ON and press Enter.
iii) Close Command Prompt and restart your PC.

and this is the mod. rom: https://15254b2dcaab7f5478ab-24461f391e20b7336331d5789078af53.ssl.cf1.rackcdn.com/ethereum.vanillaforums.com/editor/9p/vv6t6jnquwj1.zip


Now I am getting the message "miner thread hang, restart miner" and 2 of the gpus go to 0.00 mh/s before the miner restarts...

I know this is related to the OC, because probably those two cards can not manage it, but how can I solve the problem? (keeping the OC...)


Thank you for you help

[alevlaslo]

https://bitcointalk.org/index.php?topic=1610345.msg16655047#msg16655047

[adaseb]

Any idea why every 2 minutes or so the speed goes down and up with some of my rigs. The Miner Manager keeps blinking every 2-3 minutes or so that one of my rigs is running slower than usually. Usually it runs 10-20mh/s slower for 1 second and then goes back to normal.

Any way to fix this or is it just a very slow CPU ?

[osnwt]

Usually it means your CPU is doing something like Windows Update search (if you are on Windows).
As a first try you may run miner with high priority (either with start /high command or by changing the priority in the process list).
Usually all unnecessary processes should be disabled (such as Windows Update, background search indexer, etc).
Or use Linux, it just works even on 1GB RAM without issues.

[adaseb]

Usually it means your CPU is doing something like Windows Update search (if you are on Windows).
As a first try you may run miner with high priority (either with start /high command or by changing the priority in the process list).
Usually all unnecessary processes should be disabled (such as Windows Update, background search indexer, etc).
Or use Linux, it just works even on 1GB RAM without issues.

Whats the command to run Claymore on high priority?


EDIT, ahh its a Windows feature.

Never knew about it

[osnwt]

As a first try you may run miner with high priority (either with start /high command or by changing the priority in the process list).

Whats the command to run Claymore on high priority?

See above: start /high "Title" command args...
Run "start /?" on Windows command prompt for details.

With slow 2-core CPUs it helps a lot on Windows and is the easiest way.

[adaseb]

As a first try you may run miner with high priority (either with start /high command or by changing the priority in the process list).

Whats the command to run Claymore on high priority?

See above: start /high "Title" command args...
Run "start /?" on Windows command prompt for details.

With slow 2-core CPUs it helps a lot on Windows and is the easiest way.

I wish it was a 2-core CPU, its a single core Sempron.

[Yadstiker]

Usually it means your CPU is doing something like Windows Update search (if you are on Windows).
As a first try you may run miner with high priority (either with start /high command or by changing the priority in the process list).
Usually all unnecessary processes should be disabled (such as Windows Update, background search indexer, etc).
Or use Linux, it just works even on 1GB RAM without issues.

I start the miner with above normal priority. That helps quite a lot. Otherwise, the hash rate is low some times.

[osnwt]

I wish it was a 2-core CPU, its a single core Sempron.

I observed that running on 2-core CPU Win7 searches for updates and uses 100% of one core for a while. With single core it should be worse. I like Linux, the miner uses less than 6% of CPU and the system almost does not use it. So I can run a zcash CPU miner in addition to CDM :-)

[hacko.bg]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

[Claymore]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

If ethminer works, try "-eres 0" option. It must help, but note in a few epochs you will not be able to mine soon anyway, even with ethminer.

[orionpaxx12]

@claymore i added a 1070 to my rig and i can see the temp of all my other cards except the 1070 any advice, i am using the recommended drivers and everything system is running fine just no temp on the 1070, thanks for the awesome miner, hope you make one for zcash, cant wait for the launch to see if you will

[adaseb]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

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




For some reason if I only have a bunch of 2GB cards in a rig without any larger cards like 3GB-4GB then mining doesn't start either.

So if you have a rig like 2GB,2Gb,2GB,2GB,2GB

Swap one of them with 3GB/4GB and try again

3GB,2GB,2GB,2GB,2GB,2GB

[adamvp]

Any idea why every 2 minutes or so the speed goes down and up with some of my rigs. The Miner Manager keeps blinking every 2-3 minutes or so that one of my rigs is running slower than usually. Usually it runs 10-20mh/s slower for 1 second and then goes back to normal.

Any way to fix this or is it just a very slow CPU ?

In my case.. reboot helped :d I had high fluctuations beetween 24-28.5 mhs.. after reboot it was stable 28.4 at least first 20hours..

[hacko.bg]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

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




For some reason if I only have a bunch of 2GB cards in a rig without any larger cards like 3GB-4GB then mining doesn't start either.

So if you have a rig like 2GB,2Gb,2GB,2GB,2GB

Swap one of them with 3GB/4GB and try again

3GB,2GB,2GB,2GB,2GB,2GB

What you suggest is a no go, have one rig with R9 290. The 290 did start mining but because the rig has other 2GB cards, the miner restarted in the end.
What Claymore suggested did help. Obviously 2GB cards are going EOL in terms of Ethereum, although everybody said POS will come before 2GB become obsolete.
Claymore, have you considered some sort of DAG file chunking in your miner, is it doable at all? When I started mining ethereum in the beginning of the year, I've read that somebody did made it for 5xxx/6xxx cards back then, although never looked in the question.

[CoRpO]

Claymore, have you considered some sort of DAG file chunking in your miner, is it doable at all? When I started mining ethereum in the beginning of the year, I've read that somebody did made it for 5xxx/6xxx cards back then, although never looked in the question.

IIRC, that wasn't working (invalid shares due to errors in calculations)

[adamvp]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

Try other ethash aalgo please f.e Expanse

[hacko.bg]

Since last epoch #83 and a new DAG file, I'm unable to mine with my rigs with 2GB cards.
They all are with 8 cards (270, 270x, 265, etc.) and all stopped mining last night and unable to bring them online since then.
Rigs are with Ubunty LTS 14.04 x64

Variables are checked and reched, driver version is 15.302, SDK is 2.9.1 (tried different, none of them work for that matter). Miner is 5.3 to some of them, updated a few to 7.2, but still getting the dreaded "OpenCL error -4 (0) - cannot create DAG on GPU"

Rigs are with Asrock H61 Del, H81 Pro BTC some with 8GB of RAM, some with 4GB and the common is that none of them work after last night.

Tested with stock ethminer, DAG file is 1696MB and cards are currently mining, so to my observations problem is not with GPU RAM.

Any suggestions?

Try other ethash aalgo please f.e Expanse

Expanse is working as expected, problem is with DAG size.