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

[androstan1234]

It looks like the miner hangs when switching to devfee mining:

ETH - Total Speed: 402.153 Mh/s, Total Shares: 332, Rejected: 0, Time: 01:06
ETH: GPU0 29.765 Mh/s, GPU1 30.986 Mh/s, GPU2 30.675 Mh/s, GPU3 30.431 Mh/s, GPU4 29.839 Mh/s, GPU5 29.084 Mh/s, GPU6 30.319 Mh/s, GPU7 29.837 Mh/s, GPU8 30.286 Mh/s, GPU9 30.216 Mh/s, GPU10 50.664 Mh/s, GPU11 50.050 Mh/s
GPU0 t=75C fan=100%, GPU1 t=73C fan=100%, GPU2 t=72C fan=100%, GPU3 t=67C fan=100%, GPU4 t=65C fan=59%, GPU5 t=68C fan=61%, GPU6 t=72C fan=100%, GPU7 t=75C fan=100%, GPU8 t=67C fan=44%, GPU9 t=66C fan=31%; GPU10 t=64C fan=40%, GPU11 t=67C fan=45%
ETH: 06/26/18-11:37:34 - SHARE FOUND - (GPU 11) (DIFF: 9.114G)
ETH: Share accepted (109 ms)!
GPU #0: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 1:0:0)
GPU #1: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 2:0:0)
GPU #2: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 3:0:0)
GPU #3: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 4:0:0)
GPU #4: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 5:0:0)
GPU #5: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 7:0:0)
GPU #6: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 8:0:0)
GPU #7: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 9:0:0)
GPU #8: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 10:0:0)
GPU #9: Ellesmere (Radeon RX 580 Series), 8192 MB available, 36 compute units (pci bus 11:0:0)
GPU #10: GeForce GTX 1080 Ti, 11264 MB available, 28 compute units, capability: 6.1  (pci bus 12:0:0)
GPU #11: GeForce GTX 1080 Ti, 11264 MB available, 28 compute units, capability: 6.1  (pci bus 13:0:0)
ETH - Total Speed: 402.169 Mh/s, Total Shares: 333(33+22+30+21+28+23+23+25+21+21+37+49), Rejected: 0(0+0+0+0+0+0+0+0+0+0+0+0), Time: 01:06
ETH: GPU0 29.756 Mh/s, GPU1 30.996 Mh/s, GPU2 30.673 Mh/s, GPU3 30.417 Mh/s, GPU4 29.860 Mh/s, GPU5 29.126 Mh/s, GPU6 30.232 Mh/s, GPU7 30.252 Mh/s, GPU8 30.238 Mh/s, GPU9 30.201 Mh/s, GPU10 50.376 Mh/s, GPU11 50.043 Mh/s
Incorrect ETH shares: none
1 minute average ETH total speed: 402.443 Mh/s
Pool switches: ETH - 0
Current ETH share target: 0x00000000ffb34c02 (diff: 4300MH), epoch 195(2.52GB)
Max ETH share diff found: 97.799T
Current -dcri values: -dcri 8,10,8,18,20,10,8,10,8,12,30,30
GPU0 t=75C fan=100%, GPU1 t=73C fan=100%, GPU2 t=73C fan=100%, GPU3 t=67C fan=100%, GPU4 t=65C fan=59%, GPU5 t=68C fan=61%, GPU6 t=72C fan=100%, GPU7 t=74C fan=100%, GPU8 t=67C fan=44%, GPU9 t=66C fan=31%; GPU10 t=64C fan=40%, GPU11 t=67C fan=45%
ETH: 06/26/18-11:37:55 - SHARE FOUND - (GPU 5) (DIFF: 14.889G)
DevFee: ETH: Stratum - connecting to 'eth-eu.dwarfpool.com' <91.134.50.158> port 8008 (unsecure)
Minimal speed ("-minspeed" option) was set to 370 MH/s. Miner cannot reach this speed for 5 minutes, need to restart miner!
ETH: Connection lost, retry in 5 sec...
GPU #0: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 1:0:0)
GPU #1: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 2:0:0)
GPU #2: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 3:0:0)
GPU #3: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 4:0:0)
GPU #4: Ellesmere (Radeon RX 580 Series), 4096 MB available, 36 compute units (pci bus 5:0:0)
GPU #5: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 7:0:0)
GPU #6: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 8:0:0)
GPU #7: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 9:0:0)
GPU #8: Ellesmere (Radeon RX 570 Series), 4096 MB available, 32 compute units (pci bus 10:0:0)
GPU #9: Ellesmere (Radeon RX 580 Series), 8192 MB available, 36 compute units (pci bus 11:0:0)
GPU #10: GeForce GTX 1080 Ti, 11264 MB available, 28 compute units, capability: 6.1  (pci bus 12:0:0)
GPU #11: GeForce GTX 1080 Ti, 11264 MB available, 28 compute units, capability: 6.1  (pci bus 13:0:0)
ETH - Total Speed: 220.363 Mh/s, Total Shares: 334(33+22+30+21+29+25+23+25+22+23+38+50), Rejected: 0(0+0+0+0+0+0+0+0+0+0+0+0), Time: 01:25
ETH: GPU0 0.000 Mh/s, GPU1 0.000 Mh/s, GPU2 0.000 Mh/s, GPU3 0.000 Mh/s, GPU4 29.869 Mh/s, GPU5 29.087 Mh/s, GPU6 0.000 Mh/s, GPU7 0.000 Mh/s, GPU8 30.337 Mh/s, GPU9 30.200 Mh/s, GPU10 50.549 Mh/s, GPU11 50.321 Mh/s
Incorrect ETH shares: none
1 minute average ETH total speed: 220.493 Mh/s
Pool switches: ETH - 0
Max ETH share diff found: 97.799T
Current -dcri values: -dcri 8,10,8,18,20,10,8,10,8,12,30,30
DevFee: ETH: Stratum - Connected (eth-eu.dwarfpool.com:8008) (unsecure)

Any idea how to prevent this?

Remove "-minspeed" option

I'm surprised you can sleep well at night with GPUS at those temperatures.

Yeah this is what happens when summer hits.  I will probably switch over to monero soon, and use the power savings to add another 1080Ti.  Would only be slightly less profitable.

[1713493833]

1713493833

[1713493833]

1713493833

[bitgameSolo]

Out of curiosity, what kind of hash rates are you getting with those settings?

Oh, I forgot to mention the problem why I decided to ask for help. My hashrate is about 180mh on each rig. But my effective hash on pool 160-170mh. That's my problem with that. And it's not overclocking, I see that the hour of cards in rig always gets more shares per hour than some other cards. I have already perfectly adjusted the frequency of all cards, but the loss of 7-10% hasrate does not go away.

How many shares per hour do you give an average of one rig per 180Mh? I have on good rig 160 but mostly 140.

If you use the ssl stratum on ethermine you will have much fewer stale shares and therefore a higher effective hash rate.

-epool stratum+ssl://eu1.ethermine.org:5555

tried to add stratum+ssl:// to my us1 server address (where I point to ethermine network) and it didn't work. is this only for eu1 server? Also aren't peeps having issues with ssl for 11.X ?

[Vann]

For the US ethermine server you need to use the us1 for the east coast or us2 server for the west coast, whichever is closer AND the ssl stratum port 5555.

-epool stratum+ssl://us1.ethermine.org:5555

or

-epool stratum+ssl://us2.ethermine.org:5555

Everything else is the same.

[raske]

I need a little help here, I ahve some new MSI armor
they all have 8GB mem, from the GPUz is Micron mem

the stock frequencies are 1268 / 1750

I am trying to run this on 1150/2005 or 2010 but is crashing, the wattman thing is coming out andblock the PC
if I am going on lower levels to keep stzable, I cant get more than 27,5 mhs

what should I do? 

which GPU model?

RX 570, 580?

1070? 1080?

What timings are you using?  27.5 is okayish, but at mclock 2000 you should be closer to 29.5 or 30.

I forgot to tell that ther are not bios modded, I a using the compute mode on adrenaline 18.5.2
now I am testing 1155/2135, Im getiing out 28,5
Will do some omre tests with other clock combination

[androstan1234]

I need a little help here, I ahve some new MSI armor
they all have 8GB mem, from the GPUz is Micron mem

the stock frequencies are 1268 / 1750

I am trying to run this on 1150/2005 or 2010 but is crashing, the wattman thing is coming out andblock the PC
if I am going on lower levels to keep stzable, I cant get more than 27,5 mhs

what should I do? 

which GPU model?

RX 570, 580?

1070? 1080?

What timings are you using?  27.5 is okayish, but at mclock 2000 you should be closer to 29.5 or 30.

I forgot to tell that ther are not bios modded, I a using the compute mode on adrenaline 18.5.2
now I am testing 1155/2135, Im getiing out 28,5
Will do some omre tests with other clock combination

Yeah those numbers are quite acceptable for a non-biosmod.

[amortellaro]

Could someone point to a script for Windows that will query the IP and mgmt port of Claymore and return the quantity of GPUs?

I have a handful of rigs that drop a GPU after some period of mining.  With other type of crashes that occur, I have a scripted restart of the miner alone.   The only time I want to reboot the rig is if a GPU fails to be recognizes after a miner restart.

[Johnjay06]

I tried dual mining ether and xdna but xdna gives tons of rejected shares. This is on Nvidia cards and I can't figure it out. Any commands I can try? I've tried changing difficulty and that did nothing.
Fixed it, had to update the miner

[bigdaddymccarron]

A reliable source recommends 2.6gb per card, plus another 10 or 12.

@bigdaddymccarron, could it be your swap file is not big enough for the DAG files?  I'm no expert on Ubuntu, but sometimes we overlook the simplest things.

I will check that, I saw something the other day that indicated it should be 22 gb.

When I check swap usage, 0% is being used.

[mrastin]

i have problem with fan speed percentage,
claymore always shows 10% more than afterburner fan persantage,
whats the problem?
is there a bug in claymore?

[Ursul0]

Don't expect much more from it.

I have several 570's ranging from 26,5 to 28,5Mhs.

I haven't seen more then 28.5 in any 570 to date, so yours is totaly normal performance wise.

really? even my old 470 refs are doing better then the numbers you state.

rx570 can go up to 31. Normally however it hashes around 29.5-30.5
I've got a rig of rx570 4gb MSI gaming with Elpida doing mere 29.4. And the only slower 570 I've got is ASUS Expedition hashing at 28.8

[damba_]

i have problem with fan speed percentage,
claymore always shows 10% more than afterburner fan persantage,
whats the problem?
is there a bug in claymore?

I have set -tt -70 for 70% fan speed in Claymore miner, and fans are actually spinning at 80%.
It's probably something with Claymore miner, but it works like that for months.

[DeDeRa]

GPUz shows 37% fan speed, same speed its shown in MSI Afterburner or even on the competitor Phoenix Miner. But Claymore says its 51% fan speed. So yeah, Claymores readings are broken. But that isnt really a problem, the hell. Leave it as it is.

[MikeGR7]

Don't expect much more from it.

I have several 570's ranging from 26,5 to 28,5Mhs.

I haven't seen more then 28.5 in any 570 to date, so yours is totaly normal performance wise.

really? even my old 470 refs are doing better then the numbers you state.

rx570 can go up to 31. Normally however it hashes around 29.5-30.5
I've got a rig of rx570 4gb MSI gaming with Elpida doing mere 29.4. And the only slower 570 I've got is ASUS Expedition hashing at 28.8

Of course it's possible if you are willing to deal with memory errors in HWinfo.
GPUs can have hundreds of errors and seem to work fine but you have a steady higher stale shares percentage and the ocasional rejected share so i cannot accept 30mh as a 570s normal number.

[Vann]

Stale shares are shares for a block submitted after the next block is already being mined. Its result from network latency and fast block times and has nothing to do with memory errors or overclocking. Too much overclocking can definetlly cause memory errors and invalid/rejected shares though. Some of my Powercolor 4GB Red Dragon RX 570's run stable at 29.5-30.5 MH/s with a 1150 MHz core and 2075 MHz memory clock. I do have some that will only run at 1950-1975 MHz memory overclock at ~28.5 MH/s.

[DireWolfM14]

Do you mean your swap file usage is 0% while running v11.7 or v11.8?  I'm not sure exactly, but I think the DAG files are VM intensive when the miner is first starting up, so that may not be unusual.

A reliable source recommends 2.6gb per card, plus another 10 or 12.

@bigdaddymccarron, could it be your swap file is not big enough for the DAG files?  I'm no expert on Ubuntu, but sometimes we overlook the simplest things.

I will check that, I saw something the other day that indicated it should be 22 gb.

When I check swap usage, 0% is being used.

I had to revert back to v11.7 on my Linux mining rig because v11.8 keeps crashing even after reducing the overclock speeds.  On v11.7 I can run my normal overclock speeds and the miner is much more stable.  Has anyone else had any stability issues on Linux?  My 4 Windows 10 miners seem to be working fine.
<snip\>

[skyline_king]

Seems like all new version of Claymore never can reach the reported effective hashrate. so if you guys want more proficent.. use v9.8.

[jimsta]

Seems like all new version of Claymore never can reach the reported effective hashrate. so if you guys want more proficent.. use v9.8.

This is a valid point... my effective hashrate bops around alot, but i've rarely had the sustained avg hashrate to be the same or more than the reported hashrate for more than an hr.

[Ursul0]

Don't expect much more from it.

I have several 570's ranging from 26,5 to 28,5Mhs.

I haven't seen more then 28.5 in any 570 to date, so yours is totaly normal performance wise.

really? even my old 470 refs are doing better then the numbers you state.

rx570 can go up to 31. Normally however it hashes around 29.5-30.5
I've got a rig of rx570 4gb MSI gaming with Elpida doing mere 29.4. And the only slower 570 I've got is ASUS Expedition hashing at 28.8

Of course it's possible if you are willing to deal with memory errors in HWinfo.
GPUs can have hundreds of errors and seem to work fine but you have a steady higher stale shares percentage and the ocasional rejected share so i cannot accept 30mh as a 570s normal number.

also fyi mem errors in HwInfo are both those that were corrected and those that were not.

[jeremyismer]

Don't expect much more from it.

I have several 570's ranging from 26,5 to 28,5Mhs.

I haven't seen more then 28.5 in any 570 to date, so yours is totaly normal performance wise.

really? even my old 470 refs are doing better then the numbers you state.

rx570 can go up to 31. Normally however it hashes around 29.5-30.5
I've got a rig of rx570 4gb MSI gaming with Elpida doing mere 29.4. And the only slower 570 I've got is ASUS Expedition hashing at 28.8

Of course it's possible if you are willing to deal with memory errors in HWinfo.
GPUs can have hundreds of errors and seem to work fine but you have a steady higher stale shares percentage and the ocasional rejected share so i cannot accept 30mh as a 570s normal number.

https://s33.postimg.cc/hgv9pnzxb/dh31.jpg

also fyi mem errors in HwInfo are both those that were corrected and those that were not.

...sigh....I have been leading a futile existence lately. I had been using some more recent drivers (18.5.1), and from your post (and running exactly the same card) I realised I wasn't hitting the right speeds. Installed 18.2.2 and am getting the right speeds now (as a note, I had been using the compute switcher tool, but wasn't seeing any difference, the drivers I had didn't give me a compute option and I just thought Radeon Settings was being difficult, but it just wasn't there. 18.2.2 works good, 18.5.1 and some earlier versions don't. I have a 50% increase in performance now.

[mehrdadb3]

Stale shares are shares for a block submitted after the next block is already being mined. Its result from network latency and fast block times and has nothing to do with memory errors or overclocking. Too much overclocking can definetlly cause memory errors and invalid/rejected shares though. Some of my Powercolor 4GB Red Dragon RX 570's run stable at 29.5-30.5 MH/s with a 1150 MHz core and 2075 MHz memory clock. I do have some that will only run at 1950-1975 MHz memory overclock at ~28.5 MH/s.

for me too
i have msi rx 570 gaming x 4gb elpida and i cant increase memory higher than 1950 system goes to crash and i get 28.5 mh/s
and my friend have same card az me and same windows and driver but he can go memory to 2100 and get 30 mh/s
just 1 diffrent with my card and my friend is asic quality for me 76% and for he is 67%

And now, with the warming of the air, I had to decrease memory to 1900 because my system crash after 4 or 6 hours to work.when i go to 1900 its work fine but my hashrate go to 28 mh/s