AEON Mining Performance

Joel

Active Member
Jan 30, 2015
811
162
43
38
Another data point to the pile: E5-1620 V1: 650h/s in Windows. 5 threads affined to CPU 0,2,4 and HT cores 1 & 3.

The system is also running a Vega 64 loafing at 1750h/s (XMR) @ 2600rpm fan while driving a display (zero artifacts, and responsive).
 

Scott Williamson

New Member
Dec 5, 2017
6
3
3
42
Those of you who have Nvidia gear might want to try xmr-stak with the new release of CUDA 9.1. Tesla P100 went from ~1400H/s to 2650H/s on AEON after building against the new release on Ubuntu 16.04 LTS.

change device_functions.hpp to device_functions.h in /xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu and it will compile

Build it just for the GPU
Code:
cmake .. -DXMR-STAK_COMPILE=generic -DCUDA_ENABLE=ON -DOpenCL_ENABLE=OFF -DCPU_ENABLE=OFF -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.1 -DCUDA_INCLUDE_DIRS=/usr/local/cuda-9.1/include
only settings changed from stock, for the GPU, large pages enabled via the usual method.
Code:
sudo nvidia-smi -pm 1
sudo nvidia-smi -ac 715,1328
edited with some more details
 
Last edited:

marcoi

Well-Known Member
Apr 6, 2013
1,390
219
63
Gotha Florida
Those of you who have Nvidia gear might want to try xmr-stak with the new release of CUDA 9.1. Tesla P100 went from ~1400H/s to 2650H/s on AEON after building against the new release on Ubuntu 16.04 LTS.

change device_functions.hpp to device_functions.h in /xmrstak/backend/nvidia/nvcc_code/cuda_extra.cu and it will compile

Build it just for the GPU
Code:
cmake .. -DXMR-STAK_COMPILE=generic -DCUDA_ENABLE=ON -DOpenCL_ENABLE=OFF -DCPU_ENABLE=OFF -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.1 -DCUDA_INCLUDE_DIRS=/usr/local/cuda-9.1/include
only settings changed from stock, for the GPU, large pages enabled via the usual method.
Code:
sudo nvidia-smi -pm 1
sudo nvidia-smi -ac 715,1328
edited with some more details
I'm working on getting windows version compiled with cuda 9.1 so far having a few issues.
 

pyro_

Active Member
Oct 4, 2013
729
161
43
So just for laughs i fired up xmr-stak on the windows server 2016 install I just did on the new HP Microserver Gen 10 i just picked up. The AMD Opteron X3216 in it is mining Aeon at 40.6H/s

Edit: Just had a thought and change the config to only use one core and it is now mining at 78.9H/s
 

Joel

Active Member
Jan 30, 2015
811
162
43
38
I'm working on getting windows version compiled with cuda 9.1 so far having a few issues.
I'll give it a shot in the next few days with a 1060 I'm about to pick up off CL. I'll report back if the hash rate gets more interesting. Currently it seems Zcash is the way to go with NVidia cards.

If this shows promise I'll look into getting an nvidia-docker image up and running. :)
 

marcoi

Well-Known Member
Apr 6, 2013
1,390
219
63
Gotha Florida
i got a dev version of xmr-stak compiled with 9.1 so far no huge gains with 1080 card over xmrig miner i compiled with cuda 9.0.
i mining aeon though so not sure if its better with other coins.
 

keybored

Active Member
May 28, 2016
286
65
28
i got a dev version of xmr-stak compiled with 9.1 so far no huge gains with 1080 card over xmrig miner i compiled with cuda 9.0.
i mining aeon though so not sure if its better with other coins.
No huge gains, but still some gains? Or basically the same performance? 1070s should do around 1300h/s with aeon. How fast are 1080s?
 

marcoi

Well-Known Member
Apr 6, 2013
1,390
219
63
Gotha Florida
upload_2017-12-20_15-5-7.png

using the following settings so i can use the pc while mining
Code:
  // gpu: GeForce GTX 1080 architecture: 61
  //      memory: 6762/8192 MiB
  //      smx: 20
  { "index" : 0,
    "threads" : 42, "blocks" : 60,
    "bfactor" : 8, "bsleep" :  100,
    "affine_to_cpu" : false, "sync_mode" : 3,
it was about 100 h if i changed it to bfactor 8 and bsleep 25.

with xmrig-nvidia miner is was about 1225H/s with --cuda-max-threads=20 setting
 
  • Like
Reactions: JustinH