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

[delle54]

Hi there!

I've just posted a little article about my rig that runs Claymore miner. If you are interested you are welcome: Harvester 1 - The Birth of an Ether Mining Rig

I just read the article for curiosity. Nice work.

[laik2]

Hello Everyone,

I'm using Claymore's 9.6 on my rig (mix of rx 470, 480 bios moded, with 3 gpu). I was having an average Hashrate of 81.5 Mh/s (ETH+DCR) on win10. When i switched to ethOS (linux) my hashrate had a drop to 77.1 . I'm trying to figure out what's making my performance worse on linux. I really want to use a linux system basically because of stability, but i don't see this Hashrate drop as a good thing. I'm think it's probaly an issue with the driver or some wrong configuration of Claymore's software. Does anyone has any idea of what I should change or tweak in order to have a similar Hashrate that I had on win 10 ?

PS: Sorry for my bad English.

Best Regards,
Rafael.,

The theoretical windows vs linux drop is +0.4MH  for windows. Unfortunately windows drivers are a bit better than linux.

[wgt]

Hello Everyone,

I'm using Claymore's 9.6 on my rig (mix of rx 470, 480 bios moded, with 3 gpu). I was having an average Hashrate of 81.5 Mh/s (ETH+DCR) on win10. When i switched to ethOS (linux) my hashrate had a drop to 77.1 . I'm trying to figure out what's making my performance worse on linux. I really want to use a linux system basically because of stability, but i don't see this Hashrate drop as a good thing. I'm think it's probaly an issue with the driver or some wrong configuration of Claymore's software. Does anyone has any idea of what I should change or tweak in order to have a similar Hashrate that I had on win 10 ?

PS: Sorry for my bad English.

Best Regards,
Rafael.,

The theoretical windows vs linux drop is +0.4MH  for windows. Unfortunately windows drivers are a bit better than linux.

That's the point, I would take anything up to -1.0MH since ethos is pretty easy to setup and runs at least on this rig as a charm. But I'm having like -1.5Mh per gpu, wich means a lot considering i'm having 3 more gpus coming until the end of the month.

Someone told me that the overclock sweet spot on linux is different from windows, i'm testing different core/mem clocks but until now no success.

[laik2]

Hello Everyone,

I'm using Claymore's 9.6 on my rig (mix of rx 470, 480 bios moded, with 3 gpu). I was having an average Hashrate of 81.5 Mh/s (ETH+DCR) on win10. When i switched to ethOS (linux) my hashrate had a drop to 77.1 . I'm trying to figure out what's making my performance worse on linux. I really want to use a linux system basically because of stability, but i don't see this Hashrate drop as a good thing. I'm think it's probaly an issue with the driver or some wrong configuration of Claymore's software. Does anyone has any idea of what I should change or tweak in order to have a similar Hashrate that I had on win 10 ?

PS: Sorry for my bad English.

Best Regards,
Rafael.,

The theoretical windows vs linux drop is +0.4MH  for windows. Unfortunately windows drivers are a bit better than linux.

That's the point, I would take anything up to -1.0MH since ethos is pretty easy to setup and runs at least on this rig as a charm. But I'm having like -1.5Mh per gpu, wich means a lot considering i'm having 3 more gpus coming until the end of the month.

Someone told me that the overclock sweet spot on linux is different from windows, i'm testing different core/mem clocks but until now no success.

Clocking is the same as of values but result might not be as expected on windows due to Desktop loading and stuff. Besides "the lack of undervolting" linux is performing very well.

[yany]

Hi there!

I've just posted a little article about my rig that runs Claymore miner. If you are interested you are welcome: Harvester 1 - The Birth of an Ether Mining Rig

I just read the article for curiosity. Nice work.

Thanks man!

[NavyCuda]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

[Zulucia]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

You can try them and tell us the results.

[macuseri686]

@claymore If I may request a feature for an upcoming release. I myself, like many of the miners, use a bunch of headless miners connected via the network to your miner manager program. When watchdog restarts the miner, or calls the reboot.sh file, it would be really handy to have the last reboot cause recorded for easy access in the manager, or if the miner simply crashes and doesn't reboot, have the last few lines of the log file available to view in the manager. For instance "miner hangs at opencl call on GPU01". That way every single reboot doesn't warrant pulling out the monitor and connecting to the box at fault to find the issue.

Thanks in advance

[NavyCuda]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

You can try them and tell us the results.

I noticed a few other posts indicating that Vega currently doesn't work with Claymore's miner.

[rorschach1]

Problems with DUAL Mining

Hey,
when i solo mine ETH with following settings (WIN10, 6x 1070 Asus Strix): 

60% Power
+600 Memory
+0 Core

i get around 180Mh/s @ 670W

When i start ETh + SIA with same settings
i get 160 mh/s eth and ~1600 sia

is this eth drop normal? thought it would be less influence.

any improvements?
TY

Try to change -dcri parameter by push - key till acceptable values. Then change that parameter in your config/bat file. and also you have to remember that blake (sia Algorithm) is quite hot and power can be increase

[jenifive]

My Claymore is crashing DAG when switching to dev fee mining during Musicoin+DCR mining.
How to fix that? Or is there are non dev-fee version of miner around?


Why i have this error?



4x Palit 1070 Super Jet

[RuMiner]

@claymore If I may request a feature for an upcoming release. I myself, like many of the miners, use a bunch of headless miners connected via the network to your miner manager program. When watchdog restarts the miner, or calls the reboot.sh file, it would be really handy to have the last reboot cause recorded for easy access in the manager, or if the miner simply crashes and doesn't reboot, have the last few lines of the log file available to view in the manager. For instance "miner hangs at opencl call on GPU01". That way every single reboot doesn't warrant pulling out the monitor and connecting to the box at fault to find the issue.

my self-made reboot.bat sends email like tihs before reboot:

REBOOT RIG05 UPTIME 2 days 02:55:43
---------- C:\USERS\USER\DESKTOP\DUAL_MAIN\1499070431_LOG.TXT
14:20:09:412   73c   WATCHDOG: GPU 0 hangs in OpenCL call, exit
14:20:09:412   73c   WATCHDOG: GPU 0 hangs in OpenCL call, exit

Code:

Setlocal EnableDelayedExpansion
rem ############################################
set mailu=sender@gmail.com
set mailto=receiver@gmail.com
set davp=sender_password
set logdir=c:\path_to_logs
set sender="C:\Program Files\sendemail.exe"
rem ############################################
chcp 1251>nul
set "u=0"
for /f "tokens=2 delims=\" %%i in ('2^>nul reg query HKU ^| findstr /iv "classes .default"') do set /a "u+=1"
for /f "tokens=2 delims=," %%i in ('typeperf "\System\System Up Time" -sc 1 ^| findstr /rc:"\:"') do set "sec=%%i"
set "sec=%sec:"=%"
for /f "tokens=1 delims=." %%i in ("%sec%") do set "t=%%i"
set /a "ss=t%%60", "t/=60", "mm=t%%60", "t/=60", "hh=t%%24", "dd=t/24"
if %hh% lss 10 set "hh=0%hh%"
if %mm% lss 10 set "mm=0%mm%"
if %ss% lss 10 set "ss=0%ss%"
set upt=%dd% days %hh%:%mm%:%ss%
chcp 866>nul
for /f "tokens=*" %%i in (' dir /b /od "%logdir%\*.txt" ') do set "lastlog=%%i"
set tempfile=%temp%%random%
find "hangs in OpenCL call, exit" %logdir%\%lastlog% > %tempfile%
%sender% -f "%computername% <%mailu%>" -t %mailto% -u "REBOOT %computername% UPTIME %upt%" -o message-file=%tempfile%  -o message-header="X-Priority: 1" -s smtp.gmail.com -o tls=yes -xu %mailu% -xp %davp%
del %tempfile%
shutdown /r /t 5 /f

ps для pyccкoй винды cтpoкy "\System\System Up Time" зaмeнить нa "\Cиcтeмa\Bpeмя paбoты cиcтeмы"

[TechPark]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

Did you buy them for mining?

[NavyCuda]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

Did you buy them for mining?

Yes.  Mining is the excuse I used to buy them, with the hopes they'll pay for themselves, not actually turn a profit.  Will continue to game off my GTX 970 until the FE's pay for themselves.  Currently getting about 16MH from my Asus Strix GTX 970.  No OC, just whatever the card does out of the box.

[Araviel]

Getting 3mh less than excavator with 1060, any suggestions ?

[bardacuda]

First post, just joined.

Any news on support for Vega FE?  I have two of them on the way.

Did you buy them for mining?

Yes.  Mining is the excuse I used to buy them, with the hopes they'll pay for themselves, not actually turn a profit.  Will continue to game off my GTX 970 until the FE's pay for themselves.  Currently getting about 16MH from my Asus Strix GTX 970.  No OC, just whatever the card does out of the box.

I'm curious what they could do with an optimized kernel too. I would expect them to be 80% faster than Fury given the memory speed advantage, and I'm pretty sure the bus won't be a bottleneck. Last I heard though someone tried and only got low 30-something MH/s with one. Not sure which miner was used. Maybe it was sgminer.

[NavyCuda]

I'm curious what they could do with an optimized kernel too. I would expect them to be 80% faster than Fury given the memory speed advantage, and I'm pretty sure the bus won't be a bottleneck. Last I heard though someone tried and only got low 30-something MH/s with one. Not sure which miner was used. Maybe it was sgminer.

The one video I watched showed 30-35MH.  Let's assume 30MH for now.  (Yes, I saw those videos before buying two FE's, so you know the insanity is strong with me)

With using the recommended driver suggested in Claymore's miner, I saw ~2MH increase in performance with my 970.  I then dialed back the intensity so I could play old games at the same time as mining and still pull 16MH.  That's ~12%.  So if a future driver can offer a modest 12% increase on Vega's current performance, we should be able to expect 33.75MH.  I'm too new to mining to know how much the software used to mine has an effect on hashrate, but I imagine there is some optimization in the miner.

So if I can get 40MH out of each Vega FE and be at 90% GPU utilization I'll be thrilled.  If it's more than 40MH my next computer might be a threadripper with four Vega's.

I do CAD work on the side and my 970 really struggles with some models.  Even the older Quadro 5000 2GB at work beats up on it pretty hard, so Vega should be a welcome change in that regard.

[skywind912]

does anyone have a problem with their MH/S being inconsistent? i have a 6 gpu rig and it goes from anywhere 30 up to 180 mh/s am i doing something wrong? My power to my gpus are inconsistent i've changed my psu and checked all my gpus on other computers and they run fine but when i run all of them together it's really spiking like crazy....anyone have this issue?

[Gunna01]

Any idea on why a GPU will fail, then the driver crashes? It's causing my rig to BSOP after about 10 - 15 minutes of mining

[thameur]

http://

I was mining @40mh/s since today with cuda miner 2x980 untill today for 10  days but it dropped to 37 mh/s.Is this because difficulty raised ?

I heard 1060 / 1070 cards were unaffected by hashrate drop.
AMD 4xx/5xx cards also suffer from this. For every DAG Epoch, be ready for hashrate drop.

yes but there is a solution to get back hashrate to normal but with a little up temp and nothing OC  

and what's the solution?

raise core and voltage

 

I said no overclocking . it's very simple but with **temp Up** full hashrate no loss in ETH and DCR with my 4x R9 390 129.12 Mh/s ETH and 1932 Mh/s DCR but
with normal mining 119 Mh/s ETH and 1920 Mh/s DCR
I will testing this with many next DAG to confirmed the solution when all RX and all amd 2 gb to 8 gb no mining more