Topic: Wolf's XMR/BCN/DSH CPUMiner - 2x speed compared to LucasJones' - NEW 06/20/2014 - page 38. (Read 547096 times)
June 01, 2014, 12:53:01 AM
I still don't quite understand why running 3 or 7 threads on a 4 core, 8 thread CPU results in more hashes than running 1,2,4,5,6, or 8 threads
May 31, 2014, 05:19:40 PM
Is there a reason why you didn't edit out the furry images?
I had been up for four days straight reading and writing code - at that point I just didn't care
June 01, 2014, 12:19:15 AM
Win64 binaries exist but require AES-NI.
If I compile from source should I be able to run it on a processor without AES-NI then? I compiled it successfully in an Virtualbox Ubuntu 13.04 virtual machine (on a Win 7 host without AES-NI), but get:Code:
[2014-05-31 09:46:38] Using JSON-RPC 2.0
[2014-05-31 09:46:38] 2 miner threads started, using 'cryptonight' algorithm.
Illegal instruction (core dumped)
upon starting up the miner.
Configure with --disable-aes-ni
Thank you. That worked.
Without AES-NI available to me on this machine, I get a slower hash rate with your miner compared to the Lucas Jones miner. I'm sure this comes as no surprise, since you have been optimizing for AES-NI. Inside my virtual machine, per core, I get about 11 H/s with yours, and about 18.5 H/s with Lucas Jones.
May 31, 2014, 05:18:39 PM
Win64 binaries exist but require AES-NI.
If I compile from source should I be able to run it on a processor without AES-NI then? I compiled it successfully in an Virtualbox Ubuntu 13.04 virtual machine (on a Win 7 host without AES-NI), but get:Code:
[2014-05-31 09:46:38] Using JSON-RPC 2.0
[2014-05-31 09:46:38] 2 miner threads started, using 'cryptonight' algorithm.
Illegal instruction (core dumped)
upon starting up the miner.
Configure with --disable-aes-ni
May 31, 2014, 08:08:52 PM
Is there a reason why you didn't edit out the furry images?
I had been up for four days straight reading and writing code - at that point I just didn't care
grateful for your pioneering work wolfie! my mining pool hashrate reflects my hashrate on a 4770k as much faster. Consistently 200H/s+ instead of often lingering in the 100s
May 31, 2014, 04:00:20 PM
I was trying to send a donation for your work, but I guess I need payment ID.
May 31, 2014, 02:08:58 PM
Is there a reason why you didn't edit out the furry images?
'let your freak flag fly'?
i.e. be cool, don't let other people tell you how to live, and do your own thing
yeah I do have to agree, nah they weren't necessary
but like you kinda said, I'm fine with self-expression, and I might argue that that is more right than not expressing oneself
legendary
Activity: 1862
Merit: 1011
Reverse engineer from time to time
May 31, 2014, 02:05:02 PM
Is there a reason why you didn't edit out the furry images?
'let your freak flag fly'?
i.e. be cool, don't let other people tell you how to live, and do your own thing
May 31, 2014, 02:02:28 PM
Is there a reason why you didn't edit out the furry images?
'let your freak flag fly'?
i.e. be cool, don't let other people tell you how to live, and do your own thing
May 31, 2014, 02:01:46 PM
Thanks for the miner . I will try it and post my results .
legendary
Activity: 1862
Merit: 1011
Reverse engineer from time to time
May 31, 2014, 01:59:48 PM
Is there a reason why you didn't edit out the furry images?
May 31, 2014, 01:48:04 PM
got a cpuminer solution for 32 bit comp?
CryptoNight the PoW-algo is announced as 64-bit only.Maybe someone more knowledge than me, will tell you, if this true and gonna stay way (or 32-bit CryptoNight miners are possible with dirty hack or something).
May 31, 2014, 12:39:07 PM
Win64 binaries exist but require AES-NI.
If I compile from source should I be able to run it on a processor without AES-NI then? I compiled it successfully in an Virtualbox Ubuntu 13.04 virtual machine (on a Win 7 host without AES-NI), but get:Code:
[2014-05-31 09:46:38] Using JSON-RPC 2.0
[2014-05-31 09:46:38] 2 miner threads started, using 'cryptonight' algorithm.
Illegal instruction (core dumped)
upon starting up the miner.
That error is probably indicating that it is still trying to use AES-NI. Code changes might be required.
I'm not getting anywhere near the speed improvement wolf claims, but it is an improvement.
May 31, 2014, 11:55:05 AM
Win64 binaries exist but require AES-NI.
If I compile from source should I be able to run it on a processor without AES-NI then? I compiled it successfully in an Virtualbox Ubuntu 13.04 virtual machine (on a Win 7 host without AES-NI), but get:Code:
[2014-05-31 09:46:38] Using JSON-RPC 2.0
[2014-05-31 09:46:38] 2 miner threads started, using 'cryptonight' algorithm.
Illegal instruction (core dumped)
upon starting up the miner.
May 31, 2014, 11:54:50 AM
How can I set thread count?
on A8-5600 it starts 3 threads and I get a total of 42 H/s (approx 14 per thread)
At least for Intel CPUs, running (cores -1) threads gives best speedon A8-5600 it starts 3 threads and I get a total of 42 H/s (approx 14 per thread)
Otherwise with -t x, as in most cpu miners
May 31, 2014, 11:39:56 AM
got a cpuminer solution for 32 bit comp?
May 31, 2014, 10:06:37 AM
So I can just type this in terminal and it'll update?
cd ~/ && rm -rf cpuminer && mkdir cpuminer && cd cpuminer && git clone http://github.com/wolf9466/cpuminer-multi.git && pushd cpuminer-multi && cd ~/cpuminer/cpuminer-multi && ./autogen.sh && export CFLAGS="-march=native" && ./configure && make && unset CLFAGS && popd
cd ~/ && rm -rf cpuminer && mkdir cpuminer && cd cpuminer && git clone http://github.com/wolf9466/cpuminer-multi.git && pushd cpuminer-multi && cd ~/cpuminer/cpuminer-multi && ./autogen.sh && export CFLAGS="-march=native" && ./configure && make && unset CLFAGS && popd
May 31, 2014, 09:28:04 AM
Who can successfully build on OSX?
I got:
./configure: line 5666: syntax error near unexpected token `,'
./configure: line 5666: `LIBCURL_CHECK_CONFIG(, 7.15.2, ,'
when running configure
I got:
./configure: line 5666: syntax error near unexpected token `,'
./configure: line 5666: `LIBCURL_CHECK_CONFIG(, 7.15.2, ,'
when running configure
May 31, 2014, 08:06:19 AM
How can I set thread count?
on A8-5600 it starts 3 threads and I get a total of 42 H/s (approx 14 per thread)
on A8-5600 it starts 3 threads and I get a total of 42 H/s (approx 14 per thread)
May 31, 2014, 05:54:47 AM
Build from 30.05 seems to be as fast as 29-05. Maybe slightly faster. but the diff is small, like 47 per thread to 47.5 on i5-2500
Jump to: