Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

CryptoNight variant 2 migration guide #753

Closed
9 tasks done
xmrig opened this issue Sep 13, 2018 · 82 comments
Closed
9 tasks done

CryptoNight variant 2 migration guide #753

xmrig opened this issue Sep 13, 2018 · 82 comments

Comments

@xmrig
Copy link
Owner

xmrig commented Sep 13, 2018

Monero will change PoW algorithm to CryptoNight variant 2 on October 18th.

Required steps

  1. Miners and proxy should be updated to v2.8 before October 18th.
  2. variant option on each Monero pool should be set to -1 (automatic).

Notes

Test pool

http://killallasics.moneroworld.com/

Checklist

  • CPU miner is ready in dev branch.
    • v2.8 released.
  • Pure C CPU miner is ready in classic branch.
  • AMD miner ready in dev branch.
    • v2.8 released.
  • NVIDIA miner is ready in dev branch.
    • v2.8 released.
  • proxy is ready in dev branch.
    • v2.8 released.
@jetbird747
Copy link

I tested xmrig on killallasics.moneroworld.com. for cn/2. I'm seeing a 25% drop in hashrate vs cn/1. I'm using dual E5-2660 V2 cpus, is this drop in hashrate expected?

thanks

@xmrig
Copy link
Owner Author

xmrig commented Sep 19, 2018

cn/2 is slower by design, before release will be added ability to use highly optimized assembly code for Intel and AMD Ryzen CPUs, new option asm will be introduced, it will helps reduce difference in hashrate.
Thank you.

@xmrig
Copy link
Owner Author

xmrig commented Sep 25, 2018

CPU miner

CPU miner will use highly optimized Assembly code for CryptoNight variant 2, author of this code is @SChernykh. Changes already in dev branch and ready for tests. Usually no additional configuration required for it, miner will load proper ASM code automatically if it available for your CPU, for fine tuning also added global and per thread option "asm".

Possible values for "asm" option:

  • true or "auto" automatic, miner will try use proper assembly by CPUID information. This is default value.
  • false or "none" feature disabled.
  • "intel" best for Intel CPUs starting from Ivy Bridge/Sandy Bridge.
  • "ryzen" best for AMD Ryzen CPUs.

Limitations:

Only 64bit builds and single and double hash modes supported.

NVIDIA/CUDA miner

Future of NVIDIA miner now unclear, we hope @psychocrypt can create fast CUDA code for it, if not for NVIDIA can be used AMD/OpenCL miner.

AMD/OpenCL miner

  • "strided_index": 1, should be not used for CryptoNight variant 2, you will see warning message about it.
  • Added extended opencl-platform option, can be used for auto-configuration for NVIDIA GPUs.

@jetbird747
Copy link

It's my understanding that @SChernykh is still working on the double hash asm mode, hopefully this will be done and incorporated by the release date. Without it I'm seeing a 12% reduction in hashrate on my CPU system.

@SChernykh
Copy link
Contributor

@jetbird747 Double hash asm version is already integrated into xmrig.

@jetbird747
Copy link

jetbird747 commented Sep 25, 2018

@SChernykh

Using double hash I'm seeing;
CNv7: 908H/S
CNv8 873H/s

Using single hash;
CNv7: 962H/S
CNv8 846H/s

~9% reduction in hashrate

6 of 10 cores enabled on a Dual E5-2660 v2 processor; 25MB L3; Fedora 27

"threads": [
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 0 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 1 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 2 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 3 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 4 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 5 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 6 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 7 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 8 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 9 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 10 },
{ "low_power_mode" : 2, "no_prefetch" : true,  "asm" : "intel", "affine_to_cpu" : 11 }
 ]

"threads": [
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 0 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 1 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 2 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 3 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 4 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 5 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 12 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 13 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 14 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 15 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 16 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 17 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 6 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 7 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 8 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 9 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 10 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 11 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 18 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 19 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 20 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 21 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 22 },
{ "low_power_mode" : false, "no_prefetch" : true, "asm" : "intel", "affine_to_cpu" : 23 }
]

@snipeTR
Copy link

snipeTR commented Sep 25, 2018

Hardware os and xmrig info

@SChernykh
Copy link
Contributor

@jetbird747 Try to enable all 10 cores. Run single thread on 8 cores and double thread on 2 remaining cores for each CPU. This should give you the best performance.

@Bathmat
Copy link

Bathmat commented Sep 26, 2018

OS: Win10 1803
CPU: Dual E5-2670v2 (10 cores, 20 threads, 25MB L3)

CNv7: 1090 h/s
CNv8: 1055 h/s (with asm)

Interestingly, using "asm" : "intel" or "asm" : "ryzen" results in about the same hashrate. "asm" : false is only 15 h/s slower.

"threads": [
            {                "low_power_mode": 1,                "affine_to_cpu": 0            },
            {                "low_power_mode": 1,                "affine_to_cpu": 2            },
            {                "low_power_mode": 1,                "affine_to_cpu": 4            },
            {                "low_power_mode": 2,                "affine_to_cpu": 6            },
            {                "low_power_mode": 1,                "affine_to_cpu": 8            },
            {                "low_power_mode": 1,                "affine_to_cpu": 10            },
            {                "low_power_mode": 1,                "affine_to_cpu": 12            },
            {                "low_power_mode": 1,                "affine_to_cpu": 14            },
            {                "low_power_mode": 1,                "affine_to_cpu": 16            },
            {                "low_power_mode": 1,                "affine_to_cpu": 18            },
            {                "low_power_mode": 1,                "affine_to_cpu": 20            },
            {                "low_power_mode": 1,                "affine_to_cpu": 22            },
            {                "low_power_mode": 1,                "affine_to_cpu": 24            },
            {                "low_power_mode": 2,                "affine_to_cpu": 26            },
            {                "low_power_mode": 1,                "affine_to_cpu": 28            },
            {                "low_power_mode": 1,                "affine_to_cpu": 30            },
            {                "low_power_mode": 1,                "affine_to_cpu": 32            },
            {                "low_power_mode": 1,                "affine_to_cpu": 34            },
            {                "low_power_mode": 1,                "affine_to_cpu": 36            },
            {                "low_power_mode": 1,                "affine_to_cpu": 38            }
        ],

@Bathmat
Copy link

Bathmat commented Sep 26, 2018

OS: Win10 1803
GPU: GTX1050 (+160 core, +98 mem)

CNv7 (CUDA) : 325 h/s
CNv7 (OpenCL) : 319.5 h/s
CNv8 (OpenCL) : 262 h/s 👎

{            "index": 0,
            "intensity": 448,
            "worksize": 8,
            "strided_index": 0,
            "mem_chunk": 2,
            "unroll_factor": 1,
            "comp_mode": true,
            "affine_to_cpu": false        }

@SChernykh
Copy link
Contributor

CNv7 (OpenCL) : 319.5 h/s
CNv8 (OpenCL) : 262 h/

@xmrig I forgot to add my NVIDIA tuned OpenCL code for reading/writing scratchpad chunks: https://github.com/SChernykh/xmr-stak-amd/blob/master/opencl/cryptonight.cl#L728
On AMD cards it's fine to read/write to global memory directly, but on NVIDIA it's faster to first read 64 bytes to shared memory, process them and write them all back to global memory.

Can you add it? I can create a pull request in a couple of days if you don't have time right now.

@xmrig
Copy link
Owner Author

xmrig commented Sep 26, 2018

@SChernykh Yes I can, I'll do it later today.
Thank you.

@xmrig
Copy link
Owner Author

xmrig commented Sep 27, 2018

@Bathmat
Copy link

Bathmat commented Sep 27, 2018

@Bathmat @SChernykh done xmrig/xmrig-amd@2744445

Same GTX1050:

CNv7 (OpenCL) : 320 h/s
CNv8 (OpenCL) : 296 h/s 👍 Still not quite as good as AMD or CPU, but much better.

@vitaliy1985
Copy link

how to download xmrig 2.8?

@xmrig
Copy link
Owner Author

xmrig commented Sep 28, 2018

@vitaliy1985 Now it available only as source in dev branch.
Thank you.

@vitaliy1985
Copy link

"exe" will be?

@xmrig
Copy link
Owner Author

xmrig commented Sep 28, 2018

Of course it will be, but later, we still have a lot of time before October 18th for release.
Thank you.

@Bathmat
Copy link

Bathmat commented Sep 28, 2018

OS: Win10
GPU: GTX-1060 6GB (+150 core, +500 mem = 2000 core, 4300 mem)

CNv7 (CUDA): 520 h/s
CNv7 (OpenCL): 497 h/s
CNv8 (OpenCL): 446 h/s

    "threads": [
        {
            "index": 0,
            "intensity": 640,
            "worksize": 8,
            "strided_index": 0,
            "mem_chunk": 2,
            "unroll": 2,
            "comp_mode": true,
            "affine_to_cpu": false
        }
    ],

@xmrig
Copy link
Owner Author

xmrig commented Sep 30, 2018

v2.8.0-rc (release candidate) will be released on October 1 (about 24 hours from now) if no critical bugs will found, except NVIDIA miner.

https://config.xmrig.com/ is updated for v2.8 too.

Also I create Twitter account https://twitter.com/xmrig_dev it's _dev because just xmrig is already taken.

@xmrig
Copy link
Owner Author

xmrig commented Oct 1, 2018

v2.8.0-rc released.
https://github.com/xmrig/xmrig/releases/tag/v2.8.0-rc
https://github.com/xmrig/xmrig-amd/releases/tag/v2.8.0-rc
https://github.com/xmrig/xmrig-proxy/releases/tag/v2.8.0-rc

@vitaliy1985
Copy link

My xmrig.exe normally works.
What is xmrig-notls.exe for?

@2010phenix
Copy link

classic(test ~5 min each):
Last one test, strange and not detect AES and etc with VMware server:

VMware - Windows server 2008 R2 X64 (CPU E7- 4820 @ 2.00GHz - 64 core)

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=1, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://killallasics.moneroworld.com:3333
    [2018-10-08 21:11:26] Pool set diff to 100
    [2018-10-08 21:11:26] Stratum detected new block
    [2018-10-08 21:11:31] accepted: 1/1 (100.00%), 281.77 H/s at diff 100
    .....
    [2018-10-08 21:14:23] Pool set diff to 10000
    [2018-10-08 21:14:24] Stratum detected new block
    [2018-10-08 21:15:26] accepted: 134/134 (100.00%), 410.57 H/s at diff 10000
    [2018-10-08 21:15:55] accepted: 135/135 (100.00%), 410.31 H/s at diff 10000

mode. --av=1 (in config av1)
v8 asm ~414 H\s (if use key --asm auto)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=1, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://pool.supportxmr.com:3333
    [2018-10-08 21:19:30] Pool set diff to 5000
    [2018-10-08 21:19:30] Stratum detected new block
    [2018-10-08 21:19:36] accepted: 1/1 (100.00%), 545.26 H/s at diff 5000
    ....
    [2018-10-08 21:23:24] Pool set diff to 10890
    [2018-10-08 21:23:47] Stratum detected new block
    [2018-10-08 21:23:54] accepted: 13/13 (100.00%), 584.24 H/s at diff 10890

mode. --av=1 (in config av1)
Monero v7 ~584 H\s
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=3, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://killallasics.moneroworld.com:3333
    [2018-10-08 21:34:52] Pool set diff to 100
    [2018-10-08 21:34:52] Stratum detected new block
    [2018-10-08 21:34:57] accepted: 1/1 (100.00%), 226.13 H/s at diff 100
    .......
    [2018-10-08 21:39:11] Pool set diff to 10000
    [2018-10-08 21:39:11] Stratum detected new block
    [2018-10-08 21:39:31] accepted: 107/107 (100.00%), 261.39 H/s at diff 10000

mode. --av=3 (in config av auto)
v8 asm ~259 H\s (if use key --asm auto)

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=3, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://pool.supportxmr.com:3333
    [2018-10-08 21:25:15] Pool set diff to 5000
    [2018-10-08 21:25:15] Stratum detected new block
    [2018-10-08 21:25:41] accepted: 1/1 (100.00%), 290.30 H/s at diff 5000
    .....
    [2018-10-08 21:30:01] Pool set diff to 11490
    [2018-10-08 21:30:01] Stratum detected new block
    [2018-10-08 21:30:31] accepted: 15/15 (100.00%), 277.88 H/s at diff 11490

mode. --av=3 (in config auto)
Monero v7 ~283 H\s
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

very strange with AV2 double ASM

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=2, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://killallasics.moneroworld.com:3333
    [2018-10-08 21:49:23] Pool set diff to 100
    [2018-10-08 21:49:23] Stratum detected new block
    [2018-10-08 21:49:29] accepted: 1/1 (100.00%), 155.06 H/s at diff 100
    .............
    [2018-10-08 21:54:00] Pool set diff to 10000
    [2018-10-08 21:54:00] Stratum detected new block
    [2018-10-08 21:54:07] accepted: 79/79 (100.00%), 154.58 H/s at diff 10000

mode. --av=2 (in config av av2)
v8 asm ~158 H\s (if use key --asm auto)

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

  • XMRig 0.9.0-dev www.xmrig.com
  • HUGE PAGES: unavailable, disabled
  • CPU: Intel(R) Xeon(R) CPU E7- 4820 @ 2.00GHz (1)
  • CPU FEATURES: x86_64 -AES-NI
  • THREADS: 32, av=2, cryptonight/auto, donate=0%
  • STRATUM URL: stratum+tcp://pool.supportxmr.com:3333
    [2018-10-08 21:43:07] Pool set diff to 5000
    [2018-10-08 21:43:07] Stratum detected new block
    [2018-10-08 21:43:42] accepted: 1/1 (100.00%), 226.45 H/s at diff 5000
    ........
    [2018-10-08 21:47:58] Pool set diff to 5000
    [2018-10-08 21:47:58] Stratum detected new block
    [2018-10-08 21:48:01] accepted: 8/8 (100.00%), 202.32 H/s at diff 5000

double hash mode. --av=2
Monero v7 ~226 H\s

@FBI962
Copy link

FBI962 commented Oct 8, 2018

and how can i do that on Ubuntu 16.4 TLS???

@xmrig
Copy link
Owner Author

xmrig commented Oct 9, 2018

Stable v2.8.1 release (CPU/AMD/Proxy) available. For miners only minor bugfixes since v2.8.0-rc, for proxy update highly recommended to install.

@xmrig
Copy link
Owner Author

xmrig commented Oct 10, 2018

NVIDIA miner ready for cn/2 tests in dev branch https://github.com/xmrig/xmrig-nvidia/commits/dev

@FBI962
Copy link

FBI962 commented Oct 10, 2018

Stable v2.8.1 release (CPU/AMD/Proxy) available. For miners only minor bugfixes since v2.8.0-rc, for proxy update highly recommended to install.

i didnt understand

is theres any update for ubuntu 16.4 TLS??

@k0ste
Copy link

k0ste commented Oct 10, 2018

@FBI962, you should update miner at any distro.

@FBI962
Copy link

FBI962 commented Oct 10, 2018

@FBI962, you should update miner at any distro.

how can i update

what is teh command?

also is it working on ubuntu 18.4 TLS

@snipeTR
Copy link

snipeTR commented Oct 10, 2018

Read the doc

@SChernykh
Copy link
Contributor

@xmrig

NVIDIA miner ready for cn/2 tests in dev branch https://github.com/xmrig/xmrig-nvidia/commits/dev

I've optimized integer math for CUDA: SChernykh/fast_int_math_v2_cuda@2ac043c
It works faster on NVIDIA cards. I'll test optimized OpenCL version this evening.

@xmrig
Copy link
Owner Author

xmrig commented Oct 10, 2018

@SChernykh Added (you already saw it) about extra 1 H/s per card. Thank you.

@Bathmat
Copy link

Bathmat commented Oct 10, 2018

@SChernykh seeing the same as @xmrig. 1-5 h/s faster, depending on the GPU.

@SChernykh
Copy link
Contributor

@Bathmat Nice, can you post the latest numbers (CNv1 vs. CNv2) here, so I could update my performance table https://github.com/SChernykh/xmr-stak-cpu#performance ?

@Bathmat
Copy link

Bathmat commented Oct 10, 2018

@SChernykh
OS: Win7
GPU(s): GTX970, 2x GTX1050 (CUDA10, 411.7 driver)
CN/1: 483 + (2 * 300) h/s = 1083 h/s
CN/2: 455 + (2 * 291) h/s = 1037 h/s, 95.75%

Of note, this computer was slightly faster using CUDA 9.1 and 391 drivers (1100 h/s); however, I cannot get xmrig-nvidia to compile with CUDA 9.1 🤷‍♂️

For my Win10 with GTX1050:
CN/1: 325 h/s
CN/2: 318 h/s, 97.8% (that's without your latest optimizations since I haven't updated the driver for that rig yet, still using CUDA 9.1 and 397 driver)

@xmrig
Copy link
Owner Author

xmrig commented Oct 10, 2018

I cannot get xmrig-nvidia to compile with CUDA 9.1

Compile error or? -DCUDA_TOOLKIT_ROOT_DIR= allow specify path to CUDA, I recently successfully compile with CUDA 9.2 and 8.0, so no issues with 9.1 should be.

@Bathmat
Copy link

Bathmat commented Oct 10, 2018

set CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.1

Build FAILED.

"C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-nvidia.sln" (default target) (1) ->
"C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj.metaproj" (default target) (3) ->
"C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj" (default target) (4) ->
(CustomBuild target) ->
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(603): err
or : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(604): err
or : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(605): err
or : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(637): err
or : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1148): er
ror : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1594): er
ror : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(2433): er
ror : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(2433): er
ror : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(385): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xtr1common(59): error
 : class "std::enable_if<<error-constant>, int>" has no member "type" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcx
proj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(647): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(654): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(698): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(705): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(777): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(786): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(787): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(796): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(797): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(862): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xmemory0(353): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xmemory0(943): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xmemory0(1217): error
 : expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xstring(1914): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xtr1common(59): error
 : class "std::enable_if<<error-constant>, void>" has no member "type" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vc
xproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\xutility(264): error
: expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1562): er
ror : expected a ">" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\memory(1483): error :
 expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\memory(1490): error :
 expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\memory(2536): error :
 expression must have a constant value [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1562): er
ror : expected a ">" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1562): er
ror : expected a ">" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Tools\MSVC\14.15.26726\include\type_traits(1562): er
ror : expected a ">" [C:\xmrig\xmrig-nvidia-dev\cuda9.1\xmrig-cuda.vcxproj]
  CUSTOMBUILD : nvcc error : 'cicc' died with status 0xC0000005 (ACCESS_VIOLATION) [C:\xmrig\xmrig-nvidia-dev\cuda9.1\x
mrig-cuda.vcxproj]

    0 Warning(s)
    34 Error(s)

@SChernykh
Copy link
Contributor

@Bathmat It looks like CUDA 9.1 and latest Visual Studio are not really compatible.

@Bathmat
Copy link

Bathmat commented Oct 10, 2018

@SChernykh yeah, I can get 9.1 to work with stak and -vcvars_ver=14.11 and -T v141,host=x64, but that's not working with xmrig-nvidia, and I haven't had time to figure out what might work.

Also, I misspoke earlier: Drivers 391 work better than 411 on my Win7 rig. Even if I compile stak with CUDA 9.1, it's still slower with driver 411.

Oh well, I should probably just upgrade to CUDA 10.

@xmrig
Copy link
Owner Author

xmrig commented Oct 10, 2018

@Bathmat ok, then, you should use build instructions for CUDA 8 https://github.com/xmrig/xmrig-nvidia/wiki/Windows-Build-CUDA-8 but with path to CUDA 9.1.

cmake .. -G "Visual Studio 15 2017 Win64" -T v140,host=x64 -DXMRIG_DEPS=c:\xmrig-deps\msvc2015\x64 -DCUDA_TOOLKIT_ROOT_DIR="c:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v9.1"

@Bathmat
Copy link

Bathmat commented Oct 10, 2018

@xmrig thanks, that worked. Reloaded 391 drivers and speed is back up: 1100 h/s total for CNv1 and 1067 h/s for CNv2 (97%). The GTX970 stayed the same at 455 h/s for CNv2, but the 2 GTX1050s got a 15h/s boost each to 306 h/s

Also @SChernykh I tested my Win10 GTX1050 with your optimizations and CNv2 hashrate is the same; 318h/s

@Bathmat
Copy link

Bathmat commented Oct 11, 2018

Here's my last Nvidia:
OS: Win10
GPU: GTX1060 6GB (CUDA 9.1, 397 drivers), 2000 core clock, 4300 mem clock
CNv1: 520 h/s
CNv2: 512 h/s, 98.4% 🥇

EDIT: Been running it for a few hours and hashrate is stable at 507 h/s (97.5%).

@xmrig
Copy link
Owner Author

xmrig commented Oct 11, 2018

xmrig-nvidia v2.8.0-rc released https://github.com/xmrig/xmrig-nvidia/releases/tag/v2.8.0-rc

@trasherdk
Copy link

CPU....: Intel(R) Core(TM) i3-7100 CPU @ 3.90GHz
RAM....: 8 GB RAM DDR4.
OS.....: Windows 10 Pro x64 (Version 10.0.16299.611)
Driver.: Radeon Software Version Radeon Software Version Adrenalin 18.5.1

Version: xmr-stak 2.4.7 c5f0505d
Custom amt.txt config.
HashVault - CN/1

HASHRATE REPORT - CPU
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |   29.5 |   30.1 |   29.1 |  1 |   31.0 |   31.7 |   30.9 |
Totals (CPU):    60.5   61.8   60.0 H/s
-----------------------------------------------------------------
HASHRATE REPORT - AMD
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |  234.9 |  235.4 |  235.3 |  1 |  236.1 |  235.2 |  235.3 | MSI RX 550 2GB
|  2 |  238.3 |  237.3 |  237.0 |  3 |  234.5 |  236.9 |  237.0 | MSI RX 550 2GB
|  4 |  231.3 |  231.0 |  231.0 |  5 |  230.9 |  231.1 |  231.1 | GIGABYTE RX 560 4GB
|  6 |  236.8 |  237.2 |  237.3 |  7 |  238.1 |  237.9 |  237.3 | MSI RX 550 2GB
|  8 |  233.9 |  234.3 |  234.0 |  9 |  233.8 |  233.7 |  234.0 | MSI RX 550 2GB
Totals (AMD):  2348.5 2350.0 2349.3 H/s
-----------------------------------------------------------------
Totals (ALL):   2409.0 2411.8 2409.3 H/s
Highest:  2417.2 H/s
-----------------------------------------------------------------

Version: xmr-stak 2.5.0 90125127
Default autodetected amd.txt settings.
HashVault - CN/1

HASHRATE REPORT - CPU
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |   28.4 |   29.0 |   28.8 |  1 |   29.9 |   30.3 |   30.1 |
Totals (CPU):    58.3   59.2   58.9 H/s
-----------------------------------------------------------------
HASHRATE REPORT - AMD
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |  377.1 |  376.6 |  376.0 |  1 |  380.6 |  380.6 |  380.6 |
|  2 |  426.7 |  426.6 |  423.9 |  3 |  379.6 |  377.9 |  380.6 |
|  4 |  377.5 |  378.8 |  378.8 |
Totals (AMD):  1941.5 1940.5 1939.9 H/s
-----------------------------------------------------------------
Totals (ALL):   1999.8 1999.7 1998.9 H/s
Highest:  2011.0 H/s
-----------------------------------------------------------------

Version: xmr-stak 2.5.0 90125127
Custom amd.txt config. Same as ver. 2.4.7 + onroll: 8
HashVault - CN/1

HASHRATE REPORT - CPU
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |   29.7 |   29.0 |   30.5 |  1 |   31.3 |   30.4 |   31.7 |
Totals (CPU):    61.0   59.3   62.1 H/s
-----------------------------------------------------------------
HASHRATE REPORT - AMD
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |  234.5 |  235.4 |  235.2 |  1 |  236.8 |  234.9 |  235.2 | MSI RX 550 2GB
|  2 |  236.9 |  237.2 |  237.2 |  3 |  237.6 |  237.2 |  237.2 | MSI RX 550 2GB
|  4 |  230.9 |  231.1 |  231.1 |  5 |  231.3 |  231.0 |  231.0 | GIGABYTE RX 560 4GB
|  6 |  235.6 |  237.2 |  237.4 |  7 |  238.8 |  237.5 |  237.5 | MSI RX 550 2GB
|  8 |  231.4 |  233.8 |  233.9 |  9 |  235.3 |  233.6 |  233.9 | MSI RX 550 2GB
Totals (AMD):  2349.0 2348.9 2349.5 H/s
-----------------------------------------------------------------
Totals (ALL):   2410.0 2408.3 2411.7 H/s
Highest:  2417.4 H/s
-----------------------------------------------------------------

Version: xmr-stak 2.5.0 90125127
Custom amd.txt config. Same as ver. 2.4.7 + onroll: 8
KILL ALL ASICS - CN/2-2

HASHRATE REPORT - CPU
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |   29.1 |   26.2 |   28.3 |  1 |   30.5 |   27.9 |   29.7 |
Totals (CPU):    59.6   54.1   58.0 H/s
-----------------------------------------------------------------
HASHRATE REPORT - AMD
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |  198.3 |  190.6 |  189.9 |  1 |  192.4 |  191.7 |  189.9 | MSI RX 550 2GB
|  2 |  198.3 |  193.3 |  192.9 |  3 |  196.2 |  193.4 |  192.9 | MSI RX 550 2GB
|  4 |  212.3 |  211.4 |  214.0 |  5 |  217.5 |  210.5 |  214.0 | GIGABYTE RX 560 4GB
|  6 |  189.6 |  188.2 |  187.6 |  7 |  190.0 |  187.8 |  187.6 | MSI RX 550 2GB
|  8 |  202.2 |  194.9 |  193.3 |  9 |  195.2 |  195.2 |  193.3 | MSI RX 550 2GB
Totals (AMD):  1992.1 1957.1 1955.3 H/s
-----------------------------------------------------------------
Totals (ALL):   2051.7 2011.2 2013.3 H/s
Highest:  2061.2 H/s
-----------------------------------------------------------------

amd.txt

RX 550 - 2 x { "index" : 0, "intensity" : 432, "worksize" : 16, "affine_to_cpu" : false, "strided_index" : 1, "mem_chunk" : 2, "unroll" : 8, "comp_mode" : true },
RX 560 - 2 x { "index" : 2, "intensity" : 424, "worksize" : 8, "affine_to_cpu" : false, "strided_index" : 1, "mem_chunk" : 2, "unroll" : 8, "comp_mode" : true },

@trasherdk
Copy link

Just before switching back to CN/1, I copied the report on CN/2, and noticed that "Highest" is close to what I'd expect on CN/1

HASHRATE REPORT - CPU
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |   18.0 |   24.3 |   27.6 |  1 |   20.8 |   26.1 |   28.9 |
Totals (CPU):    38.8   50.4   56.5 H/s
-----------------------------------------------------------------
HASHRATE REPORT - AMD
| ID |    10s |    60s |    15m | ID |    10s |    60s |    15m |
|  0 |  179.3 |  190.6 |  191.6 |  1 |  192.4 |  190.0 |  191.7 |
|  2 |  192.4 |  192.7 |  194.4 |  3 |  190.0 |  192.6 |  194.4 |
|  4 |  214.9 |  215.3 |  212.3 |  5 |  222.8 |  216.4 |  212.2 |
|  6 |  193.1 |  189.7 |  189.8 |  7 |  185.7 |  188.8 |  189.9 |
|  8 |  199.0 |  195.3 |  194.8 |  9 |  200.5 |  195.0 |  194.8 |
Totals (AMD):  1970.0 1966.5 1966.0 H/s
-----------------------------------------------------------------
Totals (ALL):   2008.8 2016.9 2022.5 H/s
Highest:  2436.0 H/s
-----------------------------------------------------------------

@psychocrypt
Copy link

psychocrypt commented Oct 14, 2018

@trasherdk I think you reported your values to the wrong miner. This is a xmrig issue. Maybe you thought you are in fireice-uk/xmr-stak#1851

@trasherdk
Copy link

You are right. Sorry guys :)

BTW. xmrig is running just fine.

Intel(R) Xeon(R) CPU E3-1230 v5 @ 3.40GHz GenuineIntel GNU/Linux

[2018-10-14 13:57:41] accepted (12626/0) diff 7560 (37 ms)
[2018-10-14 13:57:42] accepted (12627/0) diff 7560 (83 ms)
[2018-10-14 13:57:49] accepted (12628/0) diff 7560 (37 ms)
[2018-10-14 13:57:50] speed 10s/60s/15m 260.5 261.2 259.7 H/s max 263.4 H/s

@Bathmat
Copy link

Bathmat commented Oct 14, 2018

Has anyone tested cn/2 (cnv8) with Vega and Blockchain drivers? I'm getting 1975 with cn/1 easily, but the best I can get with cn/2 is 1200 h/s (600 per thread). It's a mixed rig, and when I tried adrenaline 18.5.1, my RXs didn't want to switch to compute mode. I'll have to try again later, but just wanted to see if anyone had gotten blockchain drivers to work successfully.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

17 participants