Topic: Wolf's XMR/BCN/DSH CPUMiner - 2x speed compared to LucasJones' - NEW 06/20/2014 - page 16. (Read 547096 times)

Actually, I can get 70H/s/core on c3.8xlarge, 64H/s/core is just with this miner...

make[3]: Nothing to be done for `all-am'.
gcc -DHAVE_CONFIG_H -I.  -pthread   -Qunused-arguments -falign-loops=16 -falign-functions=16 -falign-jumps=16 -falign-labels=16  -Ofast -flto -fuse-linker-plugin -funroll-loops -fvariable-expansion-in-unroller -ftree-loop-if-convert-stores -fmerge-all-constants -fbranch-target-load-optimize2 -fsched2-use-superblocks -maes   -MT minerd-cpu-miner.o -MD -MP -MF .deps/minerd-cpu-miner.Tpo -c -o minerd-cpu-miner.o `test -f 'cpu-miner.c' || echo './'`cpu-miner.c
clang: error: unknown argument: '-falign-loops=16' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-falign-jumps=16' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-falign-labels=16' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-fuse-linker-plugin' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-fvariable-expansion-in-unroller' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-ftree-loop-if-convert-stores' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-fbranch-target-load-optimize2' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
clang: error: unknown argument: '-fsched2-use-superblocks' [-Wunused-command-line-argument-hard-error-in-future]
clang: note: this will be a hard error (cannot be downgraded to a warning) in the future
make[2]: *** [minerd-cpu-miner.o] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

Is that the 16 core Intel system or the AMD 48 core?

It's fully utilizing the cores (minus one for some reason... maybe the scheduler thread is included in the thread limit?) so I can't see a huge gain other than running it on bare metal.

The opteron cores are not as efficient per ghz as the haswells.  The only advantage is that you have physical cores without any hyperthreading.  I'll try and compile the miner on 12.04 to run it on bare metal and see what/if any advantages are seen vs running it kvm with cpu passthrough.

The server is about $3300, so it's not exactly cost effective.  Just fun to see.

edit: just noticed the instructions to install it on 12.04, derp.

It's fully utilizing the cores (minus one for some reason... maybe the scheduler thread is included in the thread limit?) so I can't see a huge gain other than running it on bare metal.

The opteron cores are not as efficient per ghz as the haswells.  The only advantage is that you have physical cores without any hyperthreading.  I'll try and compile the miner on 12.04 to run it on bare metal and see what/if any advantages are seen vs running it kvm with cpu passthrough.

The server is about $3300, so it's not exactly cost effective.  Just fun to see.

edit: just noticed the instructions to install it on 12.04, derp.

Because he has 48 physical cores, I had 16.

It was released early this morning..
https://github.com/tsiv/ccminer-cryptonight

Ok, well I guess every day I have is a "bad day" compared to the world you live in where six figures USD worth of hardware is at your disposal and you nonchalantly work them like rented mules mining XMR like it's going out of style.  Kick a brother down a Xeon server or two. will ya? 

I hit 1030 or so with 23 threads on AWS - you probably need to reduce your thread count.

Yes, 19 is a "few machines". There are plenty more to use across different pools / coins.
Difficulty is assigned by the network and will scale up and down. 444 is the starting difficulty for that pool and its gotten considerably higher now.

Lighten up, you sound like you're having a bad day

Im getting around 180H/s from my Xeon E5-2620 using this miner on Windows 7 64bit. My hashrate went up by using this miner from 150H/s. Seems bit small since Im under impression this processor should be rather good? Im new guy to CPU mining, did some mining before with my Quaddro K4000, but its just waste of time and I decided to try out with CPU.

ot: I guess you are the same wolf that runs the pool, Im seeing only 70H/s at pool statistics, normal?

1600 H/s:

4x Opteron 6344 (48 physical cores @ 2.6ghz static)
64G ram (12x4GB ddr3-
14.04, built from wolf's git source

inside KVM with cpu-passthrough (couldn't get it to run on 12.04 which is host's bare metal OS, could potentially be faster)
sysctl -w vm.nr_hugepages=144

pool reporting > 2kh, miner at 1600-1650kh/s.



Could probably get better performance by a) running it on baremetal, b) using numactl and a better task scheduler.

The cryptonight algo was not designed to be more cpu friendly than gpu friendly. It is more cpu friendly actually.
I'm not complaining, i have some cpu at home (a dual xeon 2687w and a [email protected]) - i'm not pro or against cpu mining. I have a few gpu and some cpu.
But juste mesuring the difference when mining with cpu compared to when your kids computer are not mining is not, well, a good measure.
Such a computer while mining should draw ~250W (mesured at the wall). Maybe i'm wrong, i'll let you make the measure.
A simple rig designed for mining with gpu, with a little cpu (ga2016/2020), in iddle state, draw 80w, and when mining XMR with one R9 280X, draw 250W.
With the 5 R9 280X, 2300H/s, 1000W measured at the wall.
OK, my dual xeon give me 960H/s for less power, but i think we will see a lot of optimization (for both i hope) in a near future.

Why do you think it was designed intentionally in this way ? To be fair ?
Gpu friendly coin bring gpu farm and multipool, cpu only or cpu friendly bring botnet/amazon EC2 instances (see the boolberry thread, DGA talk about 200 EC2 for himself, and he is far from being the biggest one). In both case i'm still a very little miner

EDIT :
GPU miner coming to nvidia card (not released yet)
https://bitcointalksearch.org/topic/m.7458872

First test, with 6 x 750Ti : 270W at the wall (something like 35w per card), ~160H/s per cards

You built libcurl with zlib, but you did not build a static copy; therefore it looks for one because you forced static compilation and cannot find it, thus the failure.

Pastebin config.log.

Hm... try this:

./configure is looking at curl prefix in /usr (depending on the distro). You might want configure curl like:
./configure --disable-shared --enable-static --prefix=/tmp/curl --disable-ldap --disable-sspi
make ; make install

then the miner:
./autogen.sh
./configure CFLAGS="-static' --with-libcurl=/tmp/curl
make