Pages:
Author

Topic: ethminer-0.9.41-genoil-1.1 - page 69. (Read 397360 times)

legendary
Activity: 1151
Merit: 1001
April 26, 2016, 05:38:59 AM
"feature request"

- add DAG CRC check, like in Claymore 3.3
In order to keep compatibility with generic DAG - CRC can be saved in another file in same folder as DAGs

sr. member
Activity: 438
Merit: 250
April 26, 2016, 03:31:55 AM
you should look at mean. inner mean is bugged, even before i forked it Smiley
legendary
Activity: 1151
Merit: 1001
April 26, 2016, 01:54:30 AM
Another run and
Code:
Trial 1... 14239274
Trial 2... 14239274
Trial 3... 14239274
Trial 4... 14239274
Trial 5... 14239274
min/mean/max: 14239274/14239274/14239274 H/s
inner mean: 0 H/s

:/
The card is 270X @ stock 1100/1425
legendary
Activity: 1151
Merit: 1001
April 26, 2016, 01:52:51 AM
Code:
C:\Mine\_Ether\eth105>ethminer.exe -G -M 1400000 --benchmark-warmup 10 --benchmark-tr
ial 10 --cl-local-work 128 --cl-global-work 8192
Genoil's ethminer 0.9.41-genoil-1.0.5
=====================================================================
Forked from github.com/ethereum/cpp-ethereum
CUDA kernel ported from Tim Hughes' OpenCL kernel
With contributions from RoBiK, tpruvot and sp_

Please consider a donation to:
ETH: 0xeb9310b185455f863f526dab3d245809f6854b4d

Found suitable OpenCL device [Pitcairn] with 2147483648 bytes of GPU memory
Benchmarking on platform: CL
Preparing DAG for block #1400000
  i  08:44:24|gpuminer0  workLoop 0 #00000000тAж #d6a8400bтAж
  i  08:44:24|gpuminer0  Initialising miner...
Using platfoWramr:m iAnMgD  uApc.c.e.l
erated Parallel Processing
Using device: Pitcairn(OpenCL 1.2 AMD-APP (1800.5))
Printing program log

Creating one big buffer for the DAG
Loading single big chunk kernels
Mapping one big chunk.
Creating buffer for header.
Creating mining buffer 0
Creating mining buffer 1
Trial 1... 14029873
Trial 2... 14134574
Trial 3... 14029873
Trial 4... 14029873
Trial 5... 14029873
min/mean/max: [b]14029873/14050813/14134574 H/s[/b]
inner mean: [b]4711524 [/b]H/s
Which of numbers is representative? min/mean or inner mean?
sr. member
Activity: 438
Merit: 250
April 25, 2016, 01:43:36 PM
how do i fix this ? running win7 with Microsoft Visual Studio 2015

D:\_Build\cpp-ethereum\build>cmake -DBUNDLE=miner ..
-- FindWindowsSDK: Detected Visual Studio 2012 or newer, not using the _xp tools
et variant: including SDK versions that drop XP support in search!
 - WindowsSDK dirs: C:/Program Files (x86)/Windows Kits/8.1;C:/Program Files (x8
6)/Microsoft SDKs/Windows/v8.0A;C:/Program Files (x86)/Microsoft SDKs/Windows/v8
.1A;C:/Program Files (x86)/Microsoft SDKs/Windows/v7.1A
-- ctest path: E:/Program Files (x86)/CMake/bin/ctest.exe
CMake Error at cmake/FindPackageHandleStandardArgs.cmake:138 (message):
  Could NOT find CryptoPP (missing: CRYPTOPP_INCLUDE_DIR CRYPTOPP_LIBRARY)
  (Required is at least version "5.6.2")
Call Stack (most recent call first):
  cmake/FindPackageHandleStandardArgs.cmake:374 (_FPHSA_FAILURE_MESSAGE)
  cmake/FindCryptoPP.cmake:107 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)
  cmake/EthDependencies.cmake:44 (find_package)
  CMakeLists.txt:246 (include)

this will try to make a 32-bit project. You'll ahve to specify something like cmake -DBUNDLE=miner -G "Visual Studio 14 2015 Win64" ..
full member
Activity: 196
Merit: 100
April 25, 2016, 11:58:18 AM
how do i fix this ? running win7 with Microsoft Visual Studio 2015

D:\_Build\cpp-ethereum\build>cmake -DBUNDLE=miner ..
-- FindWindowsSDK: Detected Visual Studio 2012 or newer, not using the _xp tools
et variant: including SDK versions that drop XP support in search!
 - WindowsSDK dirs: C:/Program Files (x86)/Windows Kits/8.1;C:/Program Files (x8
6)/Microsoft SDKs/Windows/v8.0A;C:/Program Files (x86)/Microsoft SDKs/Windows/v8
.1A;C:/Program Files (x86)/Microsoft SDKs/Windows/v7.1A
-- ctest path: E:/Program Files (x86)/CMake/bin/ctest.exe
CMake Error at cmake/FindPackageHandleStandardArgs.cmake:138 (message):
  Could NOT find CryptoPP (missing: CRYPTOPP_INCLUDE_DIR CRYPTOPP_LIBRARY)
  (Required is at least version "5.6.2")
Call Stack (most recent call first):
  cmake/FindPackageHandleStandardArgs.cmake:374 (_FPHSA_FAILURE_MESSAGE)
  cmake/FindCryptoPP.cmake:107 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)
  cmake/EthDependencies.cmake:44 (find_package)
  CMakeLists.txt:246 (include)
sr. member
Activity: 438
Merit: 250
April 25, 2016, 10:36:58 AM
Sad


Its above me HOW to measure the speed of this new version.
As usual,  with farm-recheck 200 its printing 5 times a second exactly same number of hashes which obviously is impossible Sad

The best you do is -G -M 1400000 --benchmark-warmup 10 --benchmark-trial 10 --cl-local-work 128 --cl-global-work 8192 (or whatever works best for you) and compare it to the 1.0.7 miner. The pipelined version ignores cl-local-work.

Anyway I think I jumped the gun with claiming an increase. I forgot to set the optimal cl-local-work on the 1.0.7 kernel.
legendary
Activity: 1151
Merit: 1001
April 25, 2016, 09:47:53 AM
Sad


Its above me HOW to measure the speed of this new version.
As usual,  with farm-recheck 200 its printing 5 times a second exactly same number of hashes which obviously is impossible Sad
sr. member
Activity: 288
Merit: 260
April 25, 2016, 09:45:19 AM
How to mine at dwarfpool in Stratum mode?!

-G -S "eth-eu.dwarfpool.com:8008/wallet/worker"  - not working

 -G -S eth-eu.dwarfpool.com:8008  -O Wallet.Worker:x - not working

:S


Genoil's ethminer doesn't support dwarfpool "stratum":

at Dwarfpools FAQ I can read,

"Low hashrate
If your calculated speed below 60% threshold over a long period of time, you send low amount of shares.
It's because of bug in some versions of ethminers. Just restart miner. Also don't use buggy Genoil v1.0.4"

are these bugs in "buggy Genoil 1.0.4" fixed at actual 1.0.7?

My miner does not even support the weird dwarfpool "stratum". I think that's what he's referring to.
legendary
Activity: 1151
Merit: 1001
April 25, 2016, 09:40:55 AM
How to mine at dwarfpool in Stratum mode?!

-G -S "eth-eu.dwarfpool.com:8008/wallet/worker"  - not working

 -G -S eth-eu.dwarfpool.com:8008  -O Wallet.Worker:x - not working

:S
sr. member
Activity: 438
Merit: 250
April 25, 2016, 08:46:50 AM
The optiminer doesn't work for me, it crashes "ethminer has stopped working" on win 7 64. Been mining for days using genoil 1.0.6, then 1.0.7

Single Sapphire Nitro 390 w/ backplate, genoil-1.0.7, following command,
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
ethminer --farm-recheck 1000 -G -S coinotron.com:3344 -O username:workerpassword --cl-local-work 256 --cl-global-work 16384, got me this,


Should i change --fram recheck to 200 or 2000? Thanks

If you use -S flag, you can safely use 2000.
newbie
Activity: 5
Merit: 0
April 25, 2016, 08:44:12 AM
The optiminer doesn't work for me, it crashes "ethminer has stopped working" on win 7 64. Been mining for days using genoil 1.0.6, then 1.0.7

Single Sapphire Nitro 390 w/ backplate, genoil-1.0.7, following command,
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
ethminer --farm-recheck 1000 -G -S coinotron.com:3344 -O username:workerpassword --cl-local-work 256 --cl-global-work 16384, got me this,
http://s26.postimg.org/ujmha77fd/n390bp.jpg

Should i change --fram recheck to 200 or 2000? Thanks
full member
Activity: 162
Merit: 100
April 25, 2016, 07:14:36 AM
genoil: its using more eletric power and doing less hashs in r9 380 cards :|
sr. member
Activity: 438
Merit: 250
April 25, 2016, 06:50:07 AM
I've got a special treat today. A guy called @optiminer from the ETH forums has open sourced his allegedly faster pipelined ethash kernels. They needed some performance tuning on my behalf, but the results are looking promising. Looking at a ~3% hashrate increase departing from my 1.0.7 on a 7950. I would appreciate if somebody could validate my results: http://forum.ethereum.org/discussion/comment/33936/#Comment_33936 (there's a win64 binary there)

No sourcecode yet. I was working on blending this in with the existing kernel but it got a bit messy so I need more time for that.



   
sr. member
Activity: 438
Merit: 250
April 25, 2016, 01:30:35 AM
at Dwarfpools FAQ I can read,

"Low hashrate
If your calculated speed below 60% threshold over a long period of time, you send low amount of shares.
It's because of bug in some versions of ethminers. Just restart miner. Also don't use buggy Genoil v1.0.4"

are these bugs in "buggy Genoil 1.0.4" fixed at actual 1.0.7?

My miner does not even support the weird dwarfpool "stratum". I think that's what he's referring to.
sr. member
Activity: 285
Merit: 250
April 25, 2016, 01:27:13 AM
Re: unstable stratum; try 1.0.7 beta in 107 branch release folder.
Re: mine on card 2 --cuda-devices 1, or mine 24/7 using --cuda-gridsize 2048
wierd, cuda devices 1, is still using both cards, so that didnt work
also using the --cuda-gridsize 2048 in the file after 'ethminer -g', made the .bat no longer open. so that also didnt work

Oh sorry should be ethminer.exe -U --cuda-grid-size 2048 or ethminer.exe -U --cuda-devices 1

The fist one will mine on both your cards with lower intensity so you can still browse, use office, etc. The second on mines with higher intensity on card 1. card 0 is your desktop card. You can play with the value of grid size until you are satisfied with desktop vs mining speed.
so it worked now, thanks! So what does lowering the grid size do? my computer is now mostly usable, and my hashrate hasn't really dropped much (it is now 39, instead of 40). Anything else besides the very small hashrate drop? I can live with that! What is the min-max range for grid size?
If anyone else is interested, I got the latest ethminer compiled with cuda 6.5 thanks to: http://cryptomining-blog.com/7759-the-latest-ethminer-0-9-41-genoil-1-0-7-for-windows/

Also for whatever reason --cuda-devices 2 never got it to see my 2nd card, I had to use -t 2
newbie
Activity: 42
Merit: 0
April 25, 2016, 12:41:02 AM
I had already used all the instructions but still my R9 270x does not work with miner
sr. member
Activity: 642
Merit: 292
April 24, 2016, 06:36:20 PM
at Dwarfpools FAQ I can read,

"Low hashrate
If your calculated speed below 60% threshold over a long period of time, you send low amount of shares.
It's because of bug in some versions of ethminers. Just restart miner. Also don't use buggy Genoil v1.0.4"

are these bugs in "buggy Genoil 1.0.4" fixed at actual 1.0.7?
sr. member
Activity: 438
Merit: 250
April 24, 2016, 04:14:30 PM
got the latest to build under windows.
but when i try again i get this
CMake Error at cmake/FindPackageHandleStandardArgs.cmake:138 (message):
  Could NOT find CryptoPP (missing: CRYPTOPP_INCLUDE_DIR CRYPTOPP_LIBRARY
  (Required is at least version "5.6.2")
Call Stack (most recent call first):
  cmake/FindPackageHandleStandardArgs.cmake:374 (_FPHSA_FAILURE_MESSAGE)
  cmake/FindCryptoPP.cmake:107 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)
  cmake/EthDependencies.cmake:44 (find_package)
  CMakeLists.txt:246 (include)

i only moved the dir to another disc. i have run getstuff.bat both with old "from when it worked" to new install of the packages

I have had this sometimes when I accidentally made a mistake in the cmake -G flag. It should be "Visual Stuidio 12 2013 Win64" if you forget Win64, you get this error...
full member
Activity: 196
Merit: 100
April 24, 2016, 10:10:57 AM
got the latest to build under windows.
but when i try again i get this
CMake Error at cmake/FindPackageHandleStandardArgs.cmake:138 (message):
  Could NOT find CryptoPP (missing: CRYPTOPP_INCLUDE_DIR CRYPTOPP_LIBRARY
  (Required is at least version "5.6.2")
Call Stack (most recent call first):
  cmake/FindPackageHandleStandardArgs.cmake:374 (_FPHSA_FAILURE_MESSAGE)
  cmake/FindCryptoPP.cmake:107 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)
  cmake/EthDependencies.cmake:44 (find_package)
  CMakeLists.txt:246 (include)

i only moved the dir to another disc. i have run getstuff.bat both with old "from when it worked" to new install of the packages
Pages:
Jump to: