JCE Miner – быстрый Cryptonote (Cryptonight) CPU+GPU майнер

JCE Miner - быстрый Cryptonote (Cryptonight) CPU+GPU майнер

Когда мы говорим о майнерах Cryptonote (Cryptonight) для графических процессоров AMD, мы предпочитаем майнер SRBMiner из-за большого числа поддерживаемых вариантов и функций, однако есть хороший конкурент, который называется JCE miner Cryptonote CPU + GPU. JCE miner – это программное обеспечение с закрытым исходным кодом, доступное в виде бинарного файла только для Windows (для графического процессора) и с платой разработчика 0,9% для графических процессоров (отдельный майнер для процессора имеет более высокую плату, для этого доступна версия для Linux). С точки зрения производительности майнер JCE в значительной степени находится на одном уровне с самыми быстрыми альтернативами, такими как SRBminer, иногда он может быть немного быстрее, иногда немного медленнее, в зависимости от алгоритма и настроек. Например, наш последний опыт показывает немного лучшую производительность для майнинга TUBE с использованием JCE по сравнению с SRB 1.7.5 на RX 580. Майнер JCE действительно поддерживает множество вариантов Cryptonothe (Cryptonight), поэтому вы можете использовать его для мой в значительной степени все популярные варианты. Существует поддержка как SSL, так и функции извлечения NiceHash.

Скачать JCE Miner

Последняя версия майнера JCE Miner доступна для скачивания по ссылкам ниже:

GitHub: DOWNLOAD JCE Miner

MEGA: DOWNLOAD JCE Miner

Features

  • GPU auto-config
  • Separate per-GPU pools and coins (to let the CPU and each GPU mine its own coin if desired)
  • Temperature and fan speed monitoring
  • Watchdog
  • Keep-alive
  • Per-GPU pause
  • Decent performance in Heavy/Haven
  • Bittube-v2 fork support
  • APU support
  • Faster and/or cached OpenCL compile

DevFee

  • 0.9% for the GPUs
  • 1.5% for the AES CPUs
  • 3.0% for the non-AES CPUs

If you mix CPU and GPU, fees are adjusted proportionally.
If you mine with GPU only, disable CPU mining with –no-cpu or manually configure zero CPU, rather than pausing your CPU to avoid paying CPU related fees for nothing, and vice-versa.

Speed

It’s hard to compare GPU Miners. There are a lot of external parameters: the card itself, its memory, the biosmod, the drivers, the overclocking, the Power Limit…

However, according to the first feedbacks, here’s the status on CN-v7:

  • JCE is faster than any other miner on small RX cards (RX550 and RX560). They are my favorite cards, and the ones I use on most of my rigs.
  • JCE is always faster than the Wolf0-based miners (Stak, Xmrig, Gateless…). This is not fair since I can read their code and they cannot read mine, but it’s a proof JCE is not a copy-paste of their code.
  • JCE is surprisingly fast on Vega, reaching 2050+ on Vega 64 on CN-v7.
  • JCE is close to SRB and Claymore otherwise, sometimes slightly above, sometimes slightly below.
  • I got mixed results on HD7800. I measured higher hashrate than other miners on my rig, but got opposite comments from some users. To be tested.
  • JCE is bad on small 1G cards compared to the legendary Claymore 9.7, but this miner is no longer supported.

On CN-v8:

  • Still always better than the open-source miners (Stak, Xmrig, Gateless…).
  • Claymore is just no longer compatible.
  • JCE is the best on older cards like HD6000 and HD7000 and R7/R9 series
  • RX560 cards reach 500+, 850+ on R9 290/X, RX570/580 cards reach 1000+
  • 420+ on HD7850 2G
  • 480+ on HD7950 3G

On CN-Heavy/HVX/Tube:

  • JCE is faster than the open-source miners on most cards:
  • 1750+ on Vega56
  • 1025+ on RX570/580 4G
  • 1200+ on RX580 8G
  • 330+ on HD7870 2G
  • 425+ on HD7950 3G

The CPU part of JCE-GPU is the exact same than the CPU-only version.

Drivers

Recommended drivers:

  • 14.4 on Windows 8.1 for HD6xxx
  • 15.12 for HD79xx
  • 18.6.1 otherwise

Конфиг:

Пример RX 480 8g Разгон 1250/2000 950mv по ядру и памяти. Выдает 1245hs на Haven.

"gpu_threads_conf" :
[
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 0, "multi_hash":944 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "gamma" : 4, "delta" : 4, "epsilon" : 4, "zeta":4, "index" : 0, "multi_hash":944 }
],

Батник:

 @echo off
set POOL=Адрес пула
set PORT=Порт пула
set WALLET=Ваш кошелек.Сложность
set PASSWORD=Пароль
 rem 0 is "automatic", you can force one of:
rem 1 = Original Cryptonight
rem 2 = Original Cryptolight
rem 3 = Cryptonight V7 fork of April-2018
rem 4 = Cryptolight V7 fork of April-2018
rem 5 = Cryptonight-Heavy
rem 6 = Cryptolight-IPBC
rem 7 = Cryptonight-XTL fork of May-2018
rem 8 = Cryptonight-Alloy
rem 9 = Cryptonight-MKT
rem 10 = Cryptonight-Arto
rem 11 = Cryptonight-Masari/Fast
rem 12 = Cryptonight-Haven
rem 13 = Cryptonight-Bittube v4 

set FORK=12

@echo on
jce_cn_gpu_miner64.exe -c CONFIG.txt --any --no-warmup --retrydelay 30 --no-cpu --keepalive --forever --low --variation %FORK% -o %POOL%:%PORT% -u %WALLET% -p %PASSWORD% %SSL% %*
@pause

–no-warmup данная функция делает хешрейт более стабильным, для Vega эту функцию лучше отключать.

Configuration

JCE provides an auto-config, with parameters:

  • –no-gpu –auto to autoconfigure CPUs and no GPU
  • –no-cpu –auto to autoconfigure GPUs and no CPU
  • –auto to autoconfigure all capable GPUs and CPUs

If you need manual config, then all rely on the configuration file. You cannot configure CPU automatically and GPU manually, or vice-versa.
The auto GPU config aims for safety and will probably be a decent but suboptimal configuration for your GPUs.

Here’s an example of a complete config with CPUs and GPUs:

"cpu_threads_conf" :
[
     { "cpu_architecture" : "auto", "affine_to_cpu" : 0, "use_cache" : true, "multi_hash":1 },
     { "cpu_architecture" : "auto", "affine_to_cpu" : 1, "use_cache" : true, "multi_hash":1 },
     { "cpu_architecture" : "auto", "affine_to_cpu" : 2, "use_cache" : true, "multi_hash":1 },
     { "cpu_architecture" : "auto", "affine_to_cpu" : 3, "use_cache" : true, "multi_hash":1 },
],

"gpu_threads_conf" :
[
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : 0, "multi_hash":464 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : 0, "multi_hash":464 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : 1, "multi_hash":208 },
     { "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : 1, "multi_hash":224 },
]

This is a real-life example from my pretty exotic and old rig:

  • The CPU part is the exact same as in JCE CPU. I used a Core2-Quad so i simply write the four normal threads.
  • The GPU part is new:

mode : can be GPU and APU, but APU is not available yet, so keep GPU

worksize : exact same meaning than in other miners, that’s the OpenCL parallelism level

the greeks : minor tuning values, in order of importance. Good values for alpha are 32, 64 or 128, and for beta 8 or 16.

index : the GPU number. Run JCE with parameter --probe to get the list. It’s in native OpenCL order, and GPU 0 is not always the main boot GPU, and may be an IGP if you have one.

multi_hash : often called intensity by other miners. Claymore parameter for it was -h. Like for CPU, that’s the number of hashes computed at the same time. While CPUs go from 1 to 6, GPUs go much higher. Often the higher the faster, but not that simple. Must be a multiple of 16.

Best known configurations

  • HD7950 (Tahiti 3GB and more) Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":464 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":464 },

Twice the same to use the double-mem mode.

  • HD7950 (Tahiti 3GB and more) Cryptonight-Heavy/Haven/Tube
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":656 },

Double-mem should be avoided here

  • HD7790 (Bonaire 1GB) Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":208 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":224 },

Two threads to use the double-mem mode, but the little 1GB VRAM doesn’t allow to use 224+224 so I use 208+224.

  • HD7870 or HD7850 (Pitcairn 2GB) Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":464 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":464 },

Twice the same to use the double-mem mode.

  • RX550 2GB Cryptonight v8
{ "mode" : "GPU", "worksize" : 16, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":432 }, 
{ "mode" : "GPU", "worksize" : 16, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":432 }, 

Twice the same to use the double-mem mode. Also try with worksize=8.

  • RX560 (Baffin 2GB) Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 8, "index" : ..., "multi_hash":464 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 128, "beta" : 8, "index" : ..., "multi_hash":464 },

Twice the same to use the double-mem mode. If a screen is plugged in the card, you may need to lower to 448 or 432

  • RX570 8G Cryptonight v7
{ "mode": "GPU", "worksize": 4, "alpha": 128, "beta": 8, "index": ..., "multi_hash": 1008 },
{ "mode": "GPU", "worksize": 4, "alpha": 128, "beta": 8, "index": ..., "multi_hash": 1008 },

Twice the same to use the double-mem mode.

  • RX580 4G Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash": 944},
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash": 944},

Twice the same to use the double-mem mode.

  • RX580 8G Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1696 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1696 },

Twice the same to use the double-mem mode.

  • RX580 8G Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1696 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1696 },

Twice the same to use the double-mem mode.

  • Vega56 Cryptonight-Heavy/Haven/Tube
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":896 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":896 },

Twice the same to use the double-mem mode.

  • RX580 8G Cryptonight-Heavy/Haven/Tube
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":832 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":832 },

Twice the same to use the double-mem mode.

  • Vega64 Cryptonight v7
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 16, "index" : ..., "multi_hash":1920 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 16, "index" : ..., "multi_hash":1920 },

or, alternative

{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1936 },
{ "mode" : "GPU", "worksize" : 8, "alpha" : 64, "beta" : 8, "index" : ..., "multi_hash":1936 },

Twice the same to use the double-mem mode.

Changelog:

0.33q CPU and 0.33b18 GPU
Quote* TurtleCoin defaults to v2
* Fix possible Masari v8 bad shares
* Waltz/Graft fork as –variation 22
0.33b17 GPU
Quote* Masari v8 support
* watchdog can restart the miner instead of quit, parameter –restart

0.33p CPU
Quote* Native Stellite v8 and Masari v8 support
0.33b16 GPU
Quote* Stellite v8
* Rig-id
* Light optim for mixed cpu/gpu mining

0.33n CPU
QuoteNew fork: Stellite v8
Rig-id parameter (optional)

0.33b15 GPU
QuoteBackport of 0.33m CPU
New fork: Turtle v2, as –variation 20
New coin: DeroGold
Remote Managment through the HTTP server
IGP autoconfig fix

0.33m CPU
QuoteNew fork: Turtle v2, as –variation 20
New coin: DeroGold
Remote Managment through the HTTP server

0.33k CPU
QuoteSeparate assembly for Bulldozer x64 on v8 algo

0.33b14 GPU
QuoteuPlexa fork, as –variation 19
Backported CPU optim from 0.33j
–legacy now keeps the CPU idle when mining only with GPU

0.33j CPU (upgrade to this one or later for CPU mining highly recommended)
QuoteuPlexa fork, as –variation 19
Better/restored speed on v8
Updated a few shitcoins forks.

0.33i CPU
QuoteBig optimization for v8, for all CPUs +2%
Sumokoin wallets default to Ryo

0.33b13 GPU
QuoteFix regression in b12 on old algos (masari, turtle, arto…)
New mode: –legacy to fix some speed regression in some card/algo combinations

0.33b11 GPU
QuoteNew coin FredCoin
Better pool-side efficiency

0.33h CPU
QuoteNew coin FredCoin
Little optim for Zen CPU on v8

0.33b10 GPU+CPU
QuoteFixed possible bad shares
Little optim for RX on v8
Little optim for Zen CPU on v8
Warning when a non-AMD gpu is enabled to mine

0.33b9 GPU (upgrade to this one or later highly recommended)
QuoteFixed Nicehash netcode
Very little optim for RX on Heavy/forks
Warmup now all automatic

0.33b8 GPU
QuoteMore stable hashrate
Speed increase in most cases

0.33b7 GPU
QuoteCryptonight-FreeHaven backported from CPU version
Cryptonight-Dark too
Fees lowered back to 0.9% for all algos
Little optims

0.33g CPU
QuoteCryptonight-FreeHaven algo with –variation 18
MoneroOcean “algo_perf” configuration possible
Updated doc

0.33f CPU
QuoteCryptolight-Dark algo with –variation 17
Parameterless config (service style)

0.33b6 GPU
QuoteMore stable hashrate
Temperature watchdog
Fixed OpenCL bug 0-2.10
Updated doc
Very light perf boost

0.33b5 GPU
QuoteCrazy huge performance boost for Heavy/HVX/Tube + 20%
Respective fees increased to 2.1%
Very light perf boost for CN-v8

0.33b4 GPU
QuoteHuge performance boost on CN-v8
Graft now defaults to CN-v8

0.33d and 0.33e CPU
QuotePerformance optimization on Non-AES-64 and v8
More coins
Fix for possible stale shares after a long pause

0.33c CPU
QuotePool-managed autoswitch (à la MoneroOcean)
Light boost on v8 algo
new assemblies: generic_fpu and generic_sse4_fpu for old CPUs
fixed AVX detect on Vishera on other recent CPUs

Рекомендуемые статьи по теме:

Добавить комментарий

Ваш адрес email не будет опубликован. Обязательные поля помечены *