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

[jeremyismer]

With the 3gb cards becoming more difficult to use I'm getting more nervous about my R7 370 4gb cards. There seems to be an issue where Claymore detects them as 3072mb on start. I've had no problems mining lately with them. But I'm getting nervous that I might.

Is the 3gb card issue because of the way Windows allocates some overhead to the memory? If so, given Windows is detecting them fine as 4gb (GPU says so) then any over head + DAG still fits within the 4gb happily that they are running fine.

Or is it more complex that than?

[john1010]

What driver version is the most compatible with this 11.6 claymore.. Any suggestion, and anyone has to try the latest amd driver is this compatible with 11.6? how about the hashing power? Thanks guy I hope i can get some help from you..

[Call_Me_Bambi]

With the 3gb cards becoming more difficult to use I'm getting more nervous about my R7 370 4gb cards. There seems to be an issue where Claymore detects them as 3072mb on start. I've had no problems mining lately with them. But I'm getting nervous that I might.

Is the 3gb card issue because of the way Windows allocates some overhead to the memory? If so, given Windows is detecting them fine as 4gb (GPU says so) then any over head + DAG still fits within the 4gb happily that they are running fine.

Or is it more complex that than?

I would not worry about what the reported memory size is if it's working OK.

A strange thing happened when I was checking Claymore's reporting of card memory size. I ran v11.5 on a 'R9 270x 4GB' and it reported 4096 MB stopped the run, then ran it on v11.6 and it reported 3072 MB stopped the run, then ran again on v11.5 and this time it reported 3072 MB. But.. after a few minutes not mining and running v11.5 again it reported 4096 MB..odd ?

[Bigdrago]

How to single mine using ethermine pool?

(Maybe this command? EthDcrMiner64.exe --epool ssl://us1.ethermine.org:5555 -esm 1 -ewal WALLET.Rigg1 - epsw x)

[big_daddy]

How to single mine using ethermine pool?

here you can solo mine ETH, https://www.ethpool.org/ , & good luck, cause you will need a lot of luck 

[Bigdrago]

How to single mine using ethermine pool?

here you can solo mine ETH, https://www.ethpool.org/ , & good luck, cause you will need a lot of luck 

Why I need good luck?
I just need the command line to start mining. I have moved from single mining on NH and moving on to ethermine.

[big_daddy]

How to single mine using ethermine pool?

here you can solo mine ETH, https://www.ethpool.org/ , & good luck, cause you will need a lot of luck 

Why I need good luck?
I just need the command line to start mining. I have moved from single mining on NH and moving on to ethermine.

sorry, I taught that you want mine solo, you mean not dual, but single mine?  so I misunderstood You, sorry 

[androstan1234]

So 825 h/sec on monero earns about as much as 29 mh/sec on eth, but monero draws 3/4 the power.  Monero seems like a no brainer for Polaris.

RX 580 is around 98 watts for 29.5mhs on Ethereum with an undervolt to around 895mv.

RX 580 is around 122 watts for 30.0mhs on Ethereum with an undervolt to around 950mv.

RX 480 can get a much lower power draw with undervolting, than the RX 580.

You would be wrong about the power draws for Ethereum.

Maybe you have some exceptional silicon?  Most of the numbers I've seen on these boards are closer to 120W at the wall for rx580, i.e.  about 750W for an ordinary 6-card rig.  Seen plenty of people pulling 800-900 watts, probably bc they're not undervolting enough.  My 580s are undervolted to around 850-860 and they pull about 115-120 from the wall while doing 29-29.5 on eth.  They only pull about 90W on monero while doing 825-850 h/sec.

In any event, even if the magnitude is off, what matters is the relative power draw between eth and xmr.  Even if the 580 is 98W for 29.5 mh/sec, xmr is better to mine at 75W for 850 h/sec.

RX 580 8GB review at 950mv undervolt - 121.8 watts for 30.05mhs

https://cryptoyeti.com/msi-rx580-armor-8g-review-and-bios-mod-guide/3/

Use, HWINFO to verify that you GPUs are accepting those undervolts.

https://www.hwinfo.com/

Because on RX Vega 56 - you cannot set an undervolt below 900mv via Adrenalin drivers.

I've confirmed that the undervolts are being accepted.

I find 659W for a 5-card rig doing 150 MH/sec believable.  I get 174 MH/sec for about 750W.

My point is that the same 5-card rig would probably do at least 4,250 h/sec on monero and pull only about 500W.

The value of the eth mined is about the same as the value of the xmr mined.

I find 98W a bit hard to believe.  I know some well optimized/tuned 470/480 rigs with good silicon can achieve something close to this, but I'm skeptical it can be done with a 580.  Maybe a freak card here or there.

Well,

That is one opinion, the record claimed by a Bitcointalk forum member is 29.3mhs for 77.5watts (Apneal) using Sapphire Nitro+ RX580 Limited Edition.

Check out his thread:

https://bitcointalk.org/index.php?topic=1881008.0

As a tip: the key is to lower the memory voltage and the GPU speed down.

He has a graph of his results, with GPU core voltages, memory voltages, memory strap used, GPU speeds and memory speeds.

Umm, that 77.5 number is the chip power, and not from the wall.  Total power is roughly two parts chip and 1 part VRM.  That's at least 115W consumed by the card.  Depending on efficiency of the PSU used, that's 120-130 watts from the wall, which is about right from my experience.

Just add on 25watts to his figures.

Za1n, on that ANN is saying he is doing, 102watts at 30mhs (down from 108watts after tinkering) with 92% efficient power supply measured at the wall.

And, 120/125 watts in dualmining Ethereum and Decred.

And, as Kilo7 says on that ANN: there are lots of RX 480 sub-100watt roms available for just mining Ethereum.

That 108 (later 102) is the power consumed by the card, not at the wall.  His 6-card rig pulls 760 from the wall, mine did 750.

760 from the wall means his cards are pulling about 120 each from the wall.  Again, that's about right.

[illusi0n_rus]

Anyone know what that means?

Code:

15:18:37:901    bcfe7700    gpu #0 dt 58.9 (31%, bad)
15:18:37:901    bcfe7700    gpu #1 dt 73.9 (39%, bad)
15:18:37:901    bcfe7700    gpu #2 dt 55.9 (29%, ok)
15:18:37:901    bcfe7700    gpu #3 dt 88.9 (47%, bad)
15:18:37:901    bcfe7700    gpu #4 dt 61.5 (65%, bad)
15:18:37:901    bcfe7700    gpu #5 dt 61.5 (65%, bad)
15:18:37:901    bcfe7700    gpu #6 dt 69.10 (36%, bad)
15:18:37:902    bcfe7700    gpu #7 dt 107.5 (114%, bad)
15:18:37:902    bcfe7700    gpu #8 dt 107.10 (54%, bad)
15:18:37:902    bcfe7700    gpu #9 dt 82.10 (43%, bad)
15:18:37:902    bcfe7700    gpu #10 dt 80.10 (41%, bad)
15:18:37:902    bcfe7700    gpu #11 dt 85.7 (59%, bad)

[MiningHelp.me]

anyone else noticed that fanmin does not work anymore with 11.6 on linux?

/clay116/ethdcrminer64 -epool $pool:port -ewal $eth_adress -epsw x -mode 1 -dbg -1 -wd 0 -fanmin 60

Works correct wit claymore 10.6 i came from before update. on 11.6 it has no effect

11.6 works normally only with >4Gb cards

[ganzocrypt]

Is already time to add new algorithms for dual mining?

which ones can be added?

[thevictimofuktyranny]

So 825 h/sec on monero earns about as much as 29 mh/sec on eth, but monero draws 3/4 the power.  Monero seems like a no brainer for Polaris.

RX 580 is around 98 watts for 29.5mhs on Ethereum with an undervolt to around 895mv.

RX 580 is around 122 watts for 30.0mhs on Ethereum with an undervolt to around 950mv.

RX 480 can get a much lower power draw with undervolting, than the RX 580.

You would be wrong about the power draws for Ethereum.

Maybe you have some exceptional silicon?  Most of the numbers I've seen on these boards are closer to 120W at the wall for rx580, i.e.  about 750W for an ordinary 6-card rig.  Seen plenty of people pulling 800-900 watts, probably bc they're not undervolting enough.  My 580s are undervolted to around 850-860 and they pull about 115-120 from the wall while doing 29-29.5 on eth.  They only pull about 90W on monero while doing 825-850 h/sec.

In any event, even if the magnitude is off, what matters is the relative power draw between eth and xmr.  Even if the 580 is 98W for 29.5 mh/sec, xmr is better to mine at 75W for 850 h/sec.

RX 580 8GB review at 950mv undervolt - 121.8 watts for 30.05mhs

https://cryptoyeti.com/msi-rx580-armor-8g-review-and-bios-mod-guide/3/

Use, HWINFO to verify that you GPUs are accepting those undervolts.

https://www.hwinfo.com/

Because on RX Vega 56 - you cannot set an undervolt below 900mv via Adrenalin drivers.

I've confirmed that the undervolts are being accepted.

I find 659W for a 5-card rig doing 150 MH/sec believable.  I get 174 MH/sec for about 750W.

My point is that the same 5-card rig would probably do at least 4,250 h/sec on monero and pull only about 500W.

The value of the eth mined is about the same as the value of the xmr mined.

I find 98W a bit hard to believe.  I know some well optimized/tuned 470/480 rigs with good silicon can achieve something close to this, but I'm skeptical it can be done with a 580.  Maybe a freak card here or there.

Well,

That is one opinion, the record claimed by a Bitcointalk forum member is 29.3mhs for 77.5watts (Apneal) using Sapphire Nitro+ RX580 Limited Edition.

Check out his thread:

https://bitcointalk.org/index.php?topic=1881008.0

As a tip: the key is to lower the memory voltage and the GPU speed down.

He has a graph of his results, with GPU core voltages, memory voltages, memory strap used, GPU speeds and memory speeds.

Umm, that 77.5 number is the chip power, and not from the wall.  Total power is roughly two parts chip and 1 part VRM.  That's at least 115W consumed by the card.  Depending on efficiency of the PSU used, that's 120-130 watts from the wall, which is about right from my experience.

Just add on 25watts to his figures.

Za1n, on that ANN is saying he is doing, 102watts at 30mhs (down from 108watts after tinkering) with 92% efficient power supply measured at the wall.

And, 120/125 watts in dualmining Ethereum and Decred.

And, as Kilo7 says on that ANN: there are lots of RX 480 sub-100watt roms available for just mining Ethereum.

That 108 (later 102) is the power consumed by the card, not at the wall.  His 6-card rig pulls 760 from the wall, mine did 750.

760 from the wall means his cards are pulling about 120 each from the wall.  Again, that's about right.

Za1n:
"My raw power usage data is:
Total rig draw: 760W (kill-a-watt reading)
PSU 92% efficient (8% inefficient):  760 - 60.8 = 699.2W (or just 760*0.92).
System idle draw (50 watts): 700 - 50 = 650W
Draw per card: 650 / 6 = ~108W"

Above are the raw stats from Za1n, before he fine-tuned his rig.

After fine tuning, Za1n drew 714watts from the wall.

On a Platinum PSU, 94% he would be at 698watts.

His system idle draw is 50 watts - when you are running mining software the system will consume more power - around 80watts from my experience.

This puts, his GPUs power draw around 105watts on his Gold certified PSU or 103 watts on a Platinium certified PSU at the wall.

Za1n, is not trying to set the lowest power consumption, he is using 900mv and 1125 GPU speed. He has headroom to reduce voltages and GPU speeds.

[rancha]

Claymore 11.5 is much more stable with lower stale shares on dual mining ETH+XVG 8gpu RX580, win10 64 rig.

[greatauror28]

Claymore 11.5 is much more stable with lower stale shares on dual mining ETH+XVG 8gpu RX580, win10 64 rig.

what total hash are you getting for ETH and what's for XVG?
Average stale shares per hour if you may.

[rancha]

Claymore 11.5 is much more stable with lower stale shares on dual mining ETH+XVG 8gpu RX580, win10 64 rig.

what total hash are you getting for ETH and what's for XVG?
Average stale shares per hour if you may.

I get 249mhs for ETH and  12465mhs for XVG. On 11.5 i get 1-2 stale share, on 11.6 i had 4-7 stales.

[androstan1234]

The victimof:  Since people do their power calculations in different ways, to compare apples and apples, it's preferable to just report power at the wall divided by total number of cards.  That's all that matters anyway for determining efficiency.  By that measure, I've never seen a rig that pulls less than ~120W/card at a decent hash rate.  Za1n's 714W result is impressive, well done.  I find that believable, but I suspect Za1n is at the outer limit of reducing power without a substantial reduction in hashrate.

And none of this actually addresses the substantive point I made, i.e. that at a given power draw (say 1440W for a typical 110V circuit), it's more profitable to mine monero than eth.

[john1010]

What driver version is the most compatible with this 11.6 claymore.. Any suggestion, and anyone has to try the latest amd driver is this compatible with 11.6? how about the hashing power? Thanks guy I hope i can get some help from you..

Guy I just want to make an followup regarding of my concern about the amd driver, has anyone of you tried this latest driver to this 11.6 version of Claymore.. Thank you in advance..

[Digital_Seytan]

Clay-Phoenix-Ethermine-Autostarter
Phoenix,Claymore Devfee no Hash Stop ,Autostart the Miner and Go Common problem with Phoenixminer and Claymore Miner is when DEVFEE comes by and the third or fourth time does not catch Shares gives these miners warning and stops the miner, to remedy this is this startbat watchdog and this ensures that the miner starts again or if necessary restart the whole miner rig, you place the contents of the start debate of the miner in this watchdog bat, and works great, you can download it at:

Download: https://github.com/digitalpara/Clay-Phoenix-Ethermine-Autostarter

Download: https://mirrorace.com/m/1qlvn

[androstan1234]

What driver version is the most compatible with this 11.6 claymore.. Any suggestion, and anyone has to try the latest amd driver is this compatible with 11.6? how about the hashing power? Thanks guy I hope i can get some help from you..

Guy I just want to make an followup regarding of my concern about the amd driver, has anyone of you tried this latest driver to this 11.6 version of Claymore.. Thank you in advance..

I have no problems using adrenalin 18.12 with claymore 11.6.  Using 580s, 570s, and a 480.

[janding]

The victimof:  Since people do their power calculations in different ways, to compare apples and apples, it's preferable to just report power at the wall divided by total number of cards.  That's all that matters anyway for determining efficiency.  By that measure, I've never seen a rig that pulls less than ~120W/card at a decent hash rate.  Za1n's 714W result is impressive, well done.  I find that believable, but I suspect Za1n is at the outer limit of reducing power without a substantial reduction in hashrate.

And none of this actually addresses the substantive point I made, i.e. that at a given power draw (say 1440W for a typical 110V circuit), it's more profitable to mine monero than eth.

Have you plugged your numbers into http://whattomine.com/ to see if that statement holds true ?