I have a mid 2010 mac pro.
It's a 2x 2.4 Xeon.
Does 330 h/s sound about right? is there anything i can do to boost it a little more.
Just saw a previous post from someone with a 2016 iMac getting around 700 h/s
The cpu is more modern but the mac pro surely wouldn't be that far behind with 8 cores?
I'm using the new xmr-stak 2.2.0
Normal HD 8gb memory, amd 5770 but it's disabled in the config because there is no guid yet to enable it, although i can't see it producing much more than 130 h/s after looking at other stats.
Comments
I get around the same on a 2x 2.66 6 core Xeon mid 2012 so that sounds good. I got about 23h/s out of a 5770... and if I increase the intensity the Mac becomes completely unresponsive. As in ctrl-z the process has no effect, no way to force quit, only option is a hard reboot.
that's great info. thanks.
Actually i thought i would look at my previous cpu config file automatically created by the wizard.
I find this a bit odd.
I have 2x Quad core Xeons so surely that means 8 cores?
My cpu file has 12, that doesn't make any sense to me.
[
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 0 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 2 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 4 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 6 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 1 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 3 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 8 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 10 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 12 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 14 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 9 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 11 },
],
That is weird. For my 12 cores I get a generated conf from 0-22 skipping all odd numbers. For whatever reason yours has included some extra odd numbers in with the even ones. What do you see in the CPU hash rate report when you type 'h' as it's running?
That's a good idea.
I get this, i compiled with 2.2.0 cpu only.
HASHRATE REPORT - CPU
| ID | 10s | 60s | 15m | ID | 10s | 60s | 15m |
| 0 | 26.5 | 28.6 | 29.1 | 1 | 27.0 | 28.7 | 29.1 |
| 2 | 26.5 | 28.7 | 29.1 | 3 | 26.6 | 28.9 | 29.1 |
| 4 | 27.0 | 28.8 | 29.1 | 5 | 26.5 | 28.5 | 29.1 |
| 6 | 27.0 | 28.8 | 29.1 | 7 | 26.5 | 28.7 | 29.1 |
| 8 | 26.5 | 28.8 | 29.1 | 9 | 26.7 | 28.9 | 29.1 |
| 10 | 27.0 | 28.9 | 29.1 | 11 | 26.7 | 28.8 | 29.1 |
Totals: 320.5 345.3 349.1 H/s
Highest: 352.4 H/s
What i would be interested to see is if i formated the drive with ubuntu to see what performance i get.
Never tried installing ubuntu on a Mac, Mac OS has allot of bells and whistles (processes), Mac hardware is very good curious if it would perform better on a simpler OS.
Are you certain you're running a machine with 8 cores? What's the output of these commands in Terminal:
sysctl -n machdep.cpu.brand_string
outputs the specific model of CPU
sysctl hw.physicalcpu
outputs the number of cores you have
Whatever the case, you're mining 12 threads. If I add the odd numbers to my config I get twice as many threads but the same hash rate.
Oh I forgot to add, I don't believe there would be much difference as far as CPU mining goes between macOS and linux, but you'd get a world of difference GPU mining.
I have 2x E5620 2.4
And I got a result of 8 which I assume is the cores?
The config file was compiled from 2.2.0 so not sure why it’s added extra lines in.
Aw I see, so it’s Mac OS that causes the gpu to run hot and the machine unresponsive, could be worth chasing with to get the gpu running also without the high risk of overheating.
Thanks again
Yeah that means you do have 8 cores, the processors have hyper threading which means each core can run 2 threads but I think it's most efficient just to run one thread per core so you might want to try deleting the odd numbered lines from the cpu setup. It's probably a bug in the cpu detection.
On other OS's you get the AMD catalyst toolkit which allows for fan monitoring and overclocking whereas on macOS it uses Apple's custom drivers which don't allow the same controls so if the machine is just being used for mining I'd definitely give it a shot.
Anyone running a 2013 Mac Pro that is mining? Looking for information on the xmr-stak-amd settings folks are using or had good feedback from?
Mac pro I have tested with xmr-stak
On macosx
Dual e5520 2.26ghz = 110+ h/s
Dual e5620 2.4ghz = 360+ h/s
Dual x5670 2.9ghz = 510+ h/s
5770 1G are slow on macosx, max getting 80h/s 2x 5770 130h/s.
But better performance on boot camp window 7.
2 x 5770 about 230h/s and single card 110h/s.
Xmr-stak CPU are slower on bootcamp win7.
I'm running Xmr-stak 2.1 on a hackintosh Ryzen and I get 275 H/s
Xmr-stak 2.2 on same Pc in Windows I get 305 H/s
This PC with macOS runs CPU related software (Cinebench, Geekbench...) faster than Windows by a bunch of % points
Where can I find a macOS compiled ver of Xmr-stak 2.2?
Are you running the xmr-stak from this website? That's effectively 2.2, it was compiled from the dev branch just before it was released as 2.2
Only a 10% decrease on macOS is not bad. It'll always be slower on macOS though because it doesn't have the AMD drivers Windows and Linux have.
thanks for your reply.
On everything else regarding the CPU the Ryzen hackintosh is faster or at par with windows, but not in mining and GPU related apps.
I have a GTX650 that works in Windows with Xmr-stack 2.2 but not on the mac.
By any chance do you think it's possible to make it work?
Thanks.
If you're just using my xmr-stak it doesn't include nvidia specific support as it requires the drivers to be installed separately, you should be able to get it working if you follow the instructions for nvidia here:
http://macminer.fabulouspanda.com/blog/index.php?post/2017/09/29/How-to-mine-Monero/XMR-on-macOS
thanks for your reply.
It actually went all fine up to some point where I got errors related to Cuda
I report then here.
The miner has been compiled but it only runs for CPU.
[ 76%] Built target xmr-stak-backend
Scanning dependencies of target xmr-stak
[ 80%] Building CXX object CMakeFiles/xmr-stak.dir/xmrstak/cli/cli-miner.cpp.o
[ 83%] Linking CXX executable bin/xmr-stak
[ 83%] Built target xmr-stak
[ 86%] Building NVCC (Device) object CMakeFiles/xmrstak_cuda_backend.dir/xmrstak/backend/nvidia/nvcc_code/xmrstak_cuda_backend_generated_cuda_extra.cu.o
nvcc fatal : The version ('90000') of the host compiler ('Apple clang') is not supported
CMake Error at xmrstak_cuda_backend_generated_cuda_extra.cu.o.Release.cmake:219 (message):
Error generating
/Users/luca/xmr-stak/CMakeFiles/xmrstak_cuda_backend.dir/xmrstak/backend/nvidia/nvcc_code/./xmrstak_cuda_backend_generated_cuda_extra.cu.o
make[2]: *** [CMakeFiles/xmrstak_cuda_backend.dir/xmrstak/backend/nvidia/nvcc_code/xmrstak_cuda_backend_generated_cuda_extra.cu.o] Error 1
make[1]: *** [CMakeFiles/xmrstak_cuda_backend.dir/all] Error 2
make: *** [all] Error 2
Ryzen:xmr-stak luca$ ./bin/xmr-stak
Please enter:
monero
Also,
I've read xmrig miner is faster (around 10%) than xmr-stack
On GitHub they have Windows versions for CPU, AMD, NVidia
Thy also give the source code and state that it can be compiled on macOS using Homebrew.
Can I compile it using as reference the link you suggested in your previous post (I'm not an expert)
Link:
https://github.com/xmrig/xmrig/releases
Thanks
Apparently there's an issue with the current Xcode developer tools that causes this error, I hadn't heard of xmrig so thanks for the link, apparently you can install it with
brew install xmrig
Thanks for your reply,
I compiled and I'm running xmrig, I get more or less the same hash rate as xmr-stack
Max 547.7 H/s on my Ryzen 7 1700 …3700Mhz (Hackintosh)
Still unabel to have the NVidia GTX 650 run.
Is there anything I can do to make it run in Cryptonight? In Windows it does 143 H/s
It looks as though that xmrig is just the cpu version and you'd have to compile separately from GitHub.
Well, I was looking to run xmr-stak on my NVidia but the compiling failed as I had a chance to comment before.
I'm running your xmr-stak with dev donation set to 0 on the CPU alone atm.
xmrig performs the same on macOS.
Any chance to have NVidia run on xmr-stak?
Thanks
It seems to be a bug relating to the version of CUDA, so you could try installing against a different version of the CUDA drivers:
https://github.com/fireice-uk/xmr-stak/issues/992