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

[ITDiver]

Hello! I'm using 4.7 version for a while, it works nice, and thank you.
Now I decide to upgrade it to 7.0(bat files the same), on one my farm I've got such error:
GPU 1, GpuMiner kx failed -48
This error occurs on each GPU, went back to 4.7

btw, what may be a reason of such error?

[rawbert]

Is 15.12 still the driver to use?  If so, where is a good place to download it?  It looks like AMD removed it from their list of downloads.

AMD page - manually select driver - (select OS and card) - "download previous drivers & software" on the right

Of course useless for newer gen. RX series...

So does 15.12 NOT work with a RX470?

What is the most stable driver version for a RX470 card then? I seem to get that my miner stops mining and says there is an issue with the nanopool.org pool. Sometimes it would just quit.

I have a Powercolor Red Devil RX470 4GB

SOmetimes it runs for 6-8 hours just fine, then other times it would just stop.

Thanks for any help!

16.8.3 is the minimum driver for the RX 470, but I always use the latest.
Edit: I meant 16.7.3.

[Q_R_V]

16.8.3 is the minimum driver for the RX 470, but I always use the latest.

Nope, 16.8.2 worked fine for my 470 4gb

[Naimon]

1. AMD cards always handle dual mode better than NVidia. Especially 290(X)/390(X) cards, they are the best. RX480 is good too. Other cards are not so good in dual mode.

but Nvidia more profitable on LBRY (1070 get 312 mh...)

Unfortunately, there is no way to get 0% drop on ETH if you mine ETH+LBC. Why? See item #2.

Fury not drop on ETH at dcri 30

On my 1070s, I've found better profitability with Claymore's ETH + LBRY than LBRY alone (only tried open source miner), as long as I max out at -dcri 500.  Actually, it's the most profitable combo I've found yet, MUCH better than ETH alone, ETH + DCR, ETH + SIA, etc.

[yun9999]

Is 15.12 still the driver to use?  If so, where is a good place to download it?  It looks like AMD removed it from their list of downloads.

AMD page - manually select driver - (select OS and card) - "download previous drivers & software" on the right

Of course useless for newer gen. RX series...

So does 15.12 NOT work with a RX470?

What is the most stable driver version for a RX470 card then? I seem to get that my miner stops mining and says there is an issue with the nanopool.org pool. Sometimes it would just quit.

I have a Powercolor Red Devil RX470 4GB

SOmetimes it runs for 6-8 hours just fine, then other times it would just stop.

Thanks for any help!

16.8.3 is the minimum driver for the RX 470, but I always use the latest.

I am using 16.7.3 so this is not true.  Certain ROMs I found 16.7.3 uses a little less wattage then 16.8.2 but not sure about 8.3 and the just release hotfix version 16.9.1


Anyone Dual mining LBRY with NVIDIA 1070?  Would be interested in knowing the good setting for it.  I'm doing 550 on Memory and -dcri of 60 and I still am only getting like 46MH.  Rumors had it at much much higher but again it may just be that, rumors.

[z0n0]

Is 15.12 still the driver to use?  If so, where is a good place to download it?  It looks like AMD removed it from their list of downloads.

AMD page - manually select driver - (select OS and card) - "download previous drivers & software" on the right

Of course useless for newer gen. RX series...

So does 15.12 NOT work with a RX470?

What is the most stable driver version for a RX470 card then? I seem to get that my miner stops mining and says there is an issue with the nanopool.org pool. Sometimes it would just quit.

I have a Powercolor Red Devil RX470 4GB

SOmetimes it runs for 6-8 hours just fine, then other times it would just stop.

Thanks for any help!

I use 16.8.3.  They are stable though the Red Devil cards don't seem to like undervolting, at all, so there they throttle ~1150MHz, secondary coin goes a little slower.  You can increase power limit though.  If there is an issue with the pool, that shouldn't affect miner, I use nanopool and never had a crash.  You can set up backup pools if you want (see OP).

15.12 does not work on any Polaris card, they need like 16.8.x or newer, RX 470 needs like 16.8.2 or 16.8.3.  15.12 is recommended for the 300, 200, 7000 series. Perhaps OP can be updated to clarify?

I use latest drivers on all my cards.
Don't see any drop in hash speed vs 15.12 on R9 cards.

[rawbert]

16.8.3 is the minimum driver for the RX 470, but I always use the latest.

Nope, 16.8.2 worked fine for my 470 4gb

Sorry, I meant 16.7.3.

[brizio71]

I have a RIG with 3 GPU RX470 e 2 GPU HD7970 and I'm running ver 7.0 o Claymore miner ETC+LBC and the hash rate change continuosly from 62 to 88 MH/s for ETC and from 95 to 112 MH/s for LBC, do you know what is the reason ?

thx

[Foss]

Hi Claymore, feature request:
set fan's speed to auto when close miner

[Claymore]

Hello! I'm using 4.7 version for a while, it works nice, and thank you.
Now I decide to upgrade it to 7.0(bat files the same), on one my farm I've got such error:
GPU 1, GpuMiner kx failed -48
This error occurs on each GPU, went back to 4.7

btw, what may be a reason of such error?

When you upgrade to new version it's a good idea to copy all files, not just .exe.

Hi Claymore, feature request:
set fan's speed to auto when close miner

Agreed, will be added in next update.

[AzzAz]

Is YIIMP supported for LBRY? My 7.0 does nor reckognize -u , -p etc

[Claymore]

Is YIIMP supported for LBRY? My 7.0 does nor reckognize -u , -p etc

"-u", "-p" are your miner options, not pool options. For options of this miner you should read OP or Readme, find sample command line for Lbry and change pool address and login/psw.

[PallWolf]

Hi, this my speed (2x480 XFX)
Version 6.3:
https://file-up.net/big_b43c7f73bd6932d57320160909122804.jpg
Version 6.4,7.0:
https://file-up.net/big_6cbf85c158863d1f7320160909123232.jpg

Speed down on 0.8 MH/s ETH and 10 MH/s on DCR per each VC, with the same others parameters.

Is it normally?

[Khad]

@claymore

Getting this with a R9 280x :

Code:

ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:04
ETH: GPU0 0.000 Mh/s
 DCR - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0
 DCR: GPU0 0.000 Mh/s
GPU0 t=50C fan=28%

What's wrong ?

[densets]

Fury X

28 eth + 47 lbc
29.6 eth + 1186 sc

252 lbc (privat miner lbc only) - most profit

https://bitcointalk.org/index.php?topic=1610345.msg16182210#msg16182210

what miner and cards are you using?

[nhan1957hacked]

1. AMD cards always handle dual mode better than NVidia. Especially 290(X)/390(X) cards, they are the best. RX480 is good too. Other cards are not so good in dual mode.

The best are Fury and Nano (29.6 eth + 1186 sc)

290x has only 29 eth + 586 sc

but Nvidia more profitable on LBRY (1070 get 312 mh, Fury get 252 mh only)

Unfortunately, there is no way to get 0% drop on ETH if you mine ETH+LBC. Why? See item #2.

Fury not drop on ETH at dcri 30

How do you get 312MH?  I'm not getting anywhere close to this number for my 1070

[Claymore]

Hi, this my speed (2x480 XFX)
Version 6.3:

Version 6.4,7.0:

Speed down on 0.8 MH/s ETH and 10 MH/s on DCR per each VC, with the same others parameters.
Is it normally?

Try different drivers, may be it's the reason, also you can try to set "-etha 2" option.

@claymore

Getting this with a R9 280x :

Code:

ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:04
ETH: GPU0 0.000 Mh/s
 DCR - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0
 DCR: GPU0 0.000 Mh/s
GPU0 t=50C fan=28%

What's wrong ?

This is not informative, the answer will be similar: something is wrong, zero hashrate is not normal.

[alfptz]

Hello!

v. 7.0 beta
Strange things: cooler speed changes periodically on some cards. The chart looks like saw. Here's the log:

16:33:43:894   328   GPU 12 temp = 60, old fan speed = 58, new fan speed = 55
16:33:43:904   328   GPU 18 temp = 59, old fan speed = 25, new fan speed = 25
16:33:43:914   328   GPU 24 temp = 61, old fan speed = 68, new fan speed = 65
16:33:46:975   328   GPU 0 temp = 67, old fan speed = 42, new fan speed = 33
16:33:47:010   328   GPU 6 temp = 67, old fan speed = 43, new fan speed = 35
16:33:47:045   328   GPU 12 temp = 60, old fan speed = 55, new fan speed = 52
16:33:47:055   328   GPU 18 temp = 59, old fan speed = 25, new fan speed = 25
16:33:47:065   328   GPU 24 temp = 61, old fan speed = 70, new fan speed = 67

Is it ok with GPU numbering?

Here's the chart.
https://drive.google.com/open?id=0B7VqAex9Ub7ncEFULTQydGJpMlk

[jstefanop]

Where the GPU fault errors fixed in recent versions of this miner for Polaris cards running the AMDGPU Pro drivers on linux...or anyone figure out whats causing them? The miner runs but these messages are spit out at like 1000/s which degrades performance and eventually fills up the dmesg log/drive with gigabytes of log files.

Code:

[  422.094330] VM fault (0x01, vmid 4) at page 135290878, read from 'TC7' (0x54433700) (132)
[  422.094350] systemd-journald[1257]: /dev/kmsg buffer overrun, some messages lost.
[  422.094434] amdgpu 0000:01:00.0: GPU fault detected: 147 0x0fd08801
[  422.094434] amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x08105FFA
[  422.094435] amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x08088001

[Claymore]

Where the GPU fault errors fixed in recent versions of this miner for Polaris cards running the AMDGPU Pro drivers on linux...or anyone figure out whats causing them? The miner runs but these messages are spit out at like 1000/s which degrades performance and eventually fills up the dmesg log/drive with gigabytes of log files.

Code:

[  422.094330] VM fault (0x01, vmid 4) at page 135290878, read from 'TC7' (0x54433700) (132)
[  422.094350] systemd-journald[1257]: /dev/kmsg buffer overrun, some messages lost.
[  422.094434] amdgpu 0000:01:00.0: GPU fault detected: 147 0x0fd08801
[  422.094434] amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x08105FFA
[  422.094435] amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x08088001

Use "-etha 2" option, available since v6.4.