ethminer-0.9.41-genoil-1.1

[restless]

But, why?!
This way you need more raM & cpu

[alfaboy23]

How much RAM does the GPU have? 1GB is not enough.
You'll want to use my fork and use the -U flag instead of -G to run in CUDA mode. CUDA doesn't need any of the setx settings.

If I'm not mistaken, this VC has only 1GB vram. But i'll still try your suggestion.

Hi can anyone verify if my set is good for eth mining?
Core i7-3770
4 GB RAM
nVidia GTX 650 vc
Windows 7 x64

No.
Foer Ether you need at least Maxwell cores (750) or newer
Notice that 760 is Kepler - a 760Ti makes 8.8MH/s atm, but its better to mine other altcoin

I see, thanks for clearing. But i'll try genoil's suggestion first and if it still didn't work out, i'll leave eth mining until i got a recommended hardware.

Thanks!

[Genoil]

If you have only 1GB of GPU RAM, there's no point trying.

[malekbaba]

Lolz even some 2gb cards are worthless for eth mining. 1Gb for eth!!!! Naaaaaaaah

[GenTarkin]

Just a heads up, I cant get -E old / all  to work on windows, the miner just crashes on start.

[user356546]

If this is not the correct place to ask this question, please let me know.

First off, big thanks to Genoil on this work; I am using the latest version with great success and better hashing than before. Thanks again!

The issue I am having is related to what I believe to be ULPS. I have a 280x and a 270x in a machine and the 280x runs fine; however, the 270x is only running at a core clock of 150 Mhz and a memory clock of the same. I have disabled ULPS in the registry and Afterburner. When I reboot and attempt to run ethminer the system bluescreens (I cannot see this occur, but on reboot it says it in the dialog box from Windows). Then I attempt to use ethminer and it works fine, but again at 150 MHz for the 270x. Also, if it matters, I cannot see the BIOS any longer when the system boots (since I added the 270x). Also, the 270x is using a USB powered riser a 750w power supply, and a fresh install of the ATI drivers and Windows.

What am I doing wrong here and what could I do to attempt to fix it? Power Limit is set to Max, PCIe link state power management is disabled, enablecrossfireautolink is set to 0, and EnableULPS is disabled in the registry. Any ideas? Thanks for the help.

EDIT: So I solved my own issue. First, I disconnected the second card to boot into the BIOS and set the primary display to PCIe. This solved the problem of not being able to see the boot screen. Then, I booted into Safemode and ran Display Driver Uninstaller from guru3d.com. Rebooted a few times and then reinstalled the latest AMD drivers. After rebooting then I reinstalled MSI Afterburner, disabled UPLS, rebooted, and it just worked. I ran ethminer and now I am happyly hashing away at the correct clock speeds. Maybe some of this will help out others, but the bottom line, uninstall the drivers THEN install your new card(s) to start mining.

[d57heinz]

Just a heads up, I cant get -E old / all  to work on windows, the miner just crashes on start.

mine looks like this on my shortcut.  "C:\Program Files\Ethereum 0.9.41\Release\ethminer.exe" --farm-recheck 200 -G -E old -F http://192.168.1.13:8546/miner/90/asrock --cl-local-work 256 --cl-global-work 16384 -t 4   

i havent seen where you need the /all  maybe i missed something.. 

Best Regards
Doug

[GenTarkin]

Just a heads up, I cant get -E old / all  to work on windows, the miner just crashes on start.

mine looks like this on my shortcut.  "C:\Program Files\Athereum 0.9.41\Release\ethminer.exe" --farm-recheck 200 -G -E old -F http://192.168.1.13:8546/miner/90/asrock --cl-local-work 256 --cl-global-work 16384 -t 4    

i havent seen where you need the /all  maybe i missed something..  

Best Regards
Doug

oh that was just a sentence structure placeholder for AND/OR =P
I mean I tried both old AND/OR all

I placed my -E old at the end of the miner arguments, it shouldnt matter but maybe it does ... actually...I know it doesnt, because I tried ethminer -E old ... without any other arguments. It still crashed on start.

[d57heinz]

Just a heads up, I cant get -E old / all  to work on windows, the miner just crashes on start.

mine looks like this on my shortcut.  "C:\Program Files\Athereum 0.9.41\Release\ethminer.exe" --farm-recheck 200 -G -E old -F http://192.168.1.13:8546/miner/90/asrock --cl-local-work 256 --cl-global-work 16384 -t 4    

i havent seen where you need the /all  maybe i missed something..  

Best Regards
Doug

oh that was just a sentence structure placeholder for AND/OR =P
I mean I tried both old AND/OR all

I placed my -E old at the end of the miner arguments, it shouldnt matter but maybe it does ... actually...I know it doesnt, because I tried ethminer -E old ... without any other arguments. It still crashed on start.

check that you have the latest version. want to say its 1.06b2 or something like that.. the first 1.06 i think had an issue where if you used the argument and there was no dags then it crashed  try it without the -E  once it creates the new dags then try with it.. sorry i cant help you further..  https://github.com/Genoil/cpp-ethereum/tree/pick-devices/releases

Best Regards
Doug

[Vaccomondus]

i need to know what farm check does exactly, it seems that my earning is increase with this set to 100

[tolazy]

ubuntu 14.04 cuda-7.5 and use cmake -DCOMPUTE=30 -DBUNDLE=cudaminer .. from cpp-ethereum/build
with: installed
apt-get install  build-essential fakeroot dh-make debconf execstack dh-modaliases dkms pkg-config
apt-get install libtool libncurses5-dev libcryptopp-dev libleveldb-dev libjsoncpp-dev libargtable2-dev libjson-rpc-cpp-dev
apt-get install libmicrohttpd-dev libreadline-dev libtool autotools-dev automake pkg-config  libssl-dev libevent-dev bsdmainutils
apt-get install -f libsdl2-dev qttools5-dev-tools libprotobuf-dev protobuf-compiler libgmp-dev cpuid


[ 65%] Building NVCC (Device) object libethash-cuda/CMakeFiles/ethash-cuda.dir/ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
cd /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir && /usr/bin/cmake -E make_directory /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//.
cd /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir && /usr/bin/cmake -D verbose:BOOL=1 -D build_configuration:STRING=Release -D generated_file:STRING=/root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o -D generated_cubin_file:STRING=/root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cubin.txt -P /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cmake
-- Removing /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
/usr/bin/cmake -E remove /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
-- Generating dependency file: /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.NVCC-depend
/usr/bin/nvcc -M -D__CUDACC__ /root/cpp-ethereum/libethash-cuda/ethash_cuda_miner_kernel.cu -o /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.NVCC-depend -ccbin /usr/bin/gcc-4.8 -m64 -DETH_JSONRPC -DETH_READLINE -DETH_MINIUPNPC -DETH_ETHASHCL -DETH_ETHASHCUDA -DETH_STRATUM -DETH_CPUID -DETH_CURL -DETH_FRONTIER -DETH_TRUE -Xcompiler ,\"-std=c++11\",\"-Wall\",\"-Wno-unknown-pragmas\",\"-Wextra\",\"-Wno-error=parentheses\",\"-pedantic\",\"-DSHAREDLIB\",\"-fPIC\",\"-fuse-ld=gold\",\"-O3\",\"-DNDEBUG\",\"-DETH_RELEASE\" --std=c++11 --disable-warnings --ptxas-options=-v -use_fast_math -lineinfo -gencode arch=compute_30,code=sm_30 -O3 -DNVCC -I/usr/include -I/root/cpp-ethereum/build2/gen -I/root/cpp-ethereum/build2 -I/root/cpp-ethereum/build2/libethash-cuda -I/usr/include -I/root/cpp-ethereum/libethash-cuda/..
nvcc fatal   : Unknown option '-std'
CMake Error at ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cmake:206 (message):
  Error generating
  /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o


make[2]: *** [libethash-cuda/CMakeFiles/ethash-cuda.dir/ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o] Error 1
make[2]: Leaving directory `/root/cpp-ethereum/build2'
make[1]: *** [libethash-cuda/CMakeFiles/ethash-cuda.dir/all] Error 2
make[1]: Leaving directory `/root/cpp-ethereum/build2'
make: *** [all] Error 2

[malandante]

Sent you 10 ETH for motivation.

Using 1.0.6 with stratum right now, seems to be the most "stable" one for my platforms.

PS put me on the private kernel list

Thanks! Much appreciated! Donators automatically get on that list, although I cant promise the kernel ever comes

I sent you 10 ETH to support your work in the miner. I wish you achieve some opencl improvement as well, count on me for any testing needed. An if you work on a special opencl kernel, count on me for the list

[Fademigo]

i need to know what farm check does exactly, it seems that my earning is increase with this set to 100

The farm check is the time interval the ethminer check for the new work. So if you decrease, the response time will be faster.

[Genoil]

1.0.6b2 from the pick-devies branch is not the latest version. It has the -E bug. That should be fixed in 1.0.6 in master branch. If you still experience issues, please send me your full command line, as well as a directory listing from your DAG storage directory.

If you look around carefully you should know I've been trying to improve the OpenCL kernel for AMD. But for now..no luck. The only result until now is a faster OpenCL kernel on Nvidia, just .5MH short of the CUDA kernel on a 970. I hope to get it at the same speed. This will be useful for Linux people, as compiling with CUDA has lead to many issues.

[malandante]

Yes I meant AMD opencl. Again if you need some testing on linux AMD rig, just drope me a line

[sorry2xs]

hello crew new to eth, i will to ask if it is possible to run 2 amd cards a 7950 and 7850 in the same rig, windows 8.1 64bit, the switches that i have read here for specifying add'l cards don't seem to work it starts to mine and stops, also i can mine with the single cards in position 0 of the mb works great but when i install the second card in position 1 it assigns itself position 0, here is my cmds i have tried,
1)single card, "ethminer.exe --farm-recheck 200 -G -F http://127.0.0.1:8332/username.workername:password --cl-global-work 8192 --cl-local-work 128: on the single 7950this works great, 20+ MHs/s
2)2cards, "ethminer.exe --farm-recheck 200 G -F http;//127.0.0.1:8332/username.workername:password  opencl-device 0,1 --cl-global-work 8192 --cl-local-128" with settings it will start to hash until the second or third "got work" then it halts and the curser blinks,also i checked the status on the amd settings and it shows that the 7950 card is operating at 99% the 7850 is at idle.
3)2cards, "ethminer.exe --farm-recheck 200 -G -F http://127.0.0.1:8332/username.workername:password" it will ether loading dag or start hashing until the second or third "got work package" and it halts.
thanks in advance

[davembg]

ubuntu 14.04 cuda-7.5 and use cmake -DCOMPUTE=30 -DBUNDLE=cudaminer .. from cpp-ethereum/build
with: installed
apt-get install  build-essential fakeroot dh-make debconf execstack dh-modaliases dkms pkg-config
apt-get install libtool libncurses5-dev libcryptopp-dev libleveldb-dev libjsoncpp-dev libargtable2-dev libjson-rpc-cpp-dev
apt-get install libmicrohttpd-dev libreadline-dev libtool autotools-dev automake pkg-config  libssl-dev libevent-dev bsdmainutils
apt-get install -f libsdl2-dev qttools5-dev-tools libprotobuf-dev protobuf-compiler libgmp-dev cpuid


[ 65%] Building NVCC (Device) object libethash-cuda/CMakeFiles/ethash-cuda.dir/ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
cd /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir && /usr/bin/cmake -E make_directory /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//.
cd /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir && /usr/bin/cmake -D verbose:BOOL=1 -D build_configuration:STRING=Release -D generated_file:STRING=/root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o -D generated_cubin_file:STRING=/root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cubin.txt -P /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cmake
-- Removing /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
/usr/bin/cmake -E remove /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o
-- Generating dependency file: /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.NVCC-depend
/usr/bin/nvcc -M -D__CUDACC__ /root/cpp-ethereum/libethash-cuda/ethash_cuda_miner_kernel.cu -o /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.NVCC-depend -ccbin /usr/bin/gcc-4.8 -m64 -DETH_JSONRPC -DETH_READLINE -DETH_MINIUPNPC -DETH_ETHASHCL -DETH_ETHASHCUDA -DETH_STRATUM -DETH_CPUID -DETH_CURL -DETH_FRONTIER -DETH_TRUE -Xcompiler ,\"-std=c++11\",\"-Wall\",\"-Wno-unknown-pragmas\",\"-Wextra\",\"-Wno-error=parentheses\",\"-pedantic\",\"-DSHAREDLIB\",\"-fPIC\",\"-fuse-ld=gold\",\"-O3\",\"-DNDEBUG\",\"-DETH_RELEASE\" --std=c++11 --disable-warnings --ptxas-options=-v -use_fast_math -lineinfo -gencode arch=compute_30,code=sm_30 -O3 -DNVCC -I/usr/include -I/root/cpp-ethereum/build2/gen -I/root/cpp-ethereum/build2 -I/root/cpp-ethereum/build2/libethash-cuda -I/usr/include -I/root/cpp-ethereum/libethash-cuda/..
nvcc fatal   : Unknown option '-std'
CMake Error at ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cmake:206 (message):
  Error generating
  /root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir//./ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o


make[2]: *** [libethash-cuda/CMakeFiles/ethash-cuda.dir/ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o] Error 1
make[2]: Leaving directory `/root/cpp-ethereum/build2'
make[1]: *** [libethash-cuda/CMakeFiles/ethash-cuda.dir/all] Error 2
make[1]: Leaving directory `/root/cpp-ethereum/build2'
make: *** [all] Error 2

edit this file:
/root/cpp-ethereum/build2/libethash-cuda/CMakeFiles/ethash-cuda.dir/ethash-cuda_generated_ethash_cuda_miner_kernel.cu.o.cmake

look for lines that set the nvcc_flags:
set(CUDA_NVCC_FLAGS ..........

-std isn't a valid option for some versions of mvcc - so, remove that option.
My line looks like this:
set(CUDA_NVCC_FLAGS --disable-warnings;--ptxas-options=-v;-use_fast_math;-lineinfo;-gencode;arch=compute_20,code=sm_20;-gencode;arch=compute_30,code=sm_30;-gencode;arch=compute_35,code=sm_35;-gencode;arch=compute_50,code=sm_50 ;; )

Also, stop building as root:)
sudo is your friend, my friend.

[davembg]

i need to know what farm check does exactly, it seems that my earning is increase with this set to 100

farm-recheck is the time (in milliseconds) interval your miner checks the farm for new work/block changes.
You will punish your farm, but, get new work sooner after a block change.

Won't make a world of difference, but, it will help a little.

Notes:
If you use a stratum proxy, a lower farm-recheck is fine as your actually just punishing your proxy.
The proxy "subscribes" to block changes.

Genoil's hard work implementing stratum should make the farm-recheck a thing of the past with stratum-capable pools.

[davembg]

hello crew new to eth, i will to ask if it is possible to run 2 amd cards a 7950 and 7850 in the same rig, windows 8.1 64bit, the switches that i have read here for specifying add'l cards don't seem to work it starts to mine and stops, also i can mine with the single cards in position 0 of the mb works great but when i install the second card in position 1 it assigns itself position 0, here is my cmds i have tried,
1)single card, "ethminer.exe --farm-recheck 200 -G -F http://127.0.0.1:8332/username.workername:password --cl-global-work 8192 --cl-local-work 128: on the single 7950this works great, 20+ MHs/s
2)2cards, "ethminer.exe --farm-recheck 200 G -F http;//127.0.0.1:8332/username.workername:password  opencl-device 0,1 --cl-global-work 8192 --cl-local-128" with settings it will start to hash until the second or third "got work" then it halts and the curser blinks,also i checked the status on the amd settings and it shows that the 7950 card is operating at 99% the 7850 is at idle.
3)2cards, "ethminer.exe --farm-recheck 200 -G -F http://127.0.0.1:8332/username.workername:password" it will ether loading dag or start hashing until the second or third "got work package" and it halts.
thanks in advance

Yeah, stop using Windows:)

Here is an example "hillbilly winders" startup bat file:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100

cd C:\Users\dave\Desktop\ethminer-0.9.41-genoil-1.0.5
ethminer -G -R . --farm-recheck 400 --cl-local-work 128 --cl-global-work 8192 -F http://ethereum.usa.nicehash.com:3500/n1c3-1HtkaiWKeKZ6MJgag6a3fUQo3BxcT2gAKv.nvda/20

There are a lot of things to check.

8GB of RAM is a good idea.
Make sure you have available disk space - DAG files are ~3GB in total.
Make sure your are using the latest AMD drivers - Crimson has been working well for most.
Take all that malware off your "Winders" box too - strip it BARE, or even do a clean install of a 64bit OS.

[sorry2xs]

thanks for the response, I have exceeded all the minimum requirements will try tomorrow, is the bat file in the proxy or in the ethminer i am noob at this, this pc was 2013 btc gpu miner has clean instal of windows as of 2 hours ago  
Edit: i am well versed in DMG, Install shield and install wizard