Rigel

This is a cryptocurrency miner for Nvidia GPUs

Supported algorithms

Dual mining

• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+alephium
• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+karlsenhash
• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+pyrinhash
• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+pyrinhashv2
• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+sha256ton
• (abelian/autolykos2/ethash/ethash/ethashb3/fishhash/octopus)+sha512256d
• karlsenhashv2+pyrinhashv2
• karlsenhashv2+sha256ton
• any single or dual algorithm combination + zil

Features

• Available on Linux and Windows
• Terminal user interface
• Failover pool support
• Temperature, clocks, fan monitoring
• Built-in GPU overclocking support with an option to apply different settings for ZIL
• Temperature limits and fan control
• HTTP API

Usage

  -a, --algorithm <ALGORITHM>
          Selects the mining algorithm

          Currently supported:
          abelian       (ABEL)
          alephium      (ALPH)
          autolykos2    (ERG, BLOCX)
          etchash       (ETC)
          ethash        (ETHW, XPB, OCTA, etc.)
          ethashb3      (HYP)
          fishhash      (IRON)
          karlsenhash   (pre-fork KLS, NXL)
          karlsenhashv2 (KLS)
          kawpow        (RVN, AIPG, XNA, CLORE, NEOX, etc.)
          nexapow       (NEXA)
          octopus       (CFX)
          pyrinhash     (PYI)
          sha256ton     (GRAM)
          sha512256d    (RXD)
          xelishash     (pre-fork XEL)
          xelishashv2   (XEL)
          zil           (ZIL)

          To dual or triple mine pass the algorithm list
          separated by `+`, e.g. `autolykos2+sha256ton+zil`

      --a1 <ALGORITHM>
          Sets the primary algorithm (alternative to `-a`)

      --a2 <ALGORITHM>
          Sets the second algorithm (alternative to `-a`)

      --a3 <ALGORITHM>
          Sets the third algorithm (alternative to `-a`)

      --coin <COIN>
          Sets the coin ticker

          Only applicable to DAG based algorithms used by multiple coins.
          Unless you're mining to a profit switching pool like Nicehash,
          setting the coin ticker has the benefit of eliminating DAG rebuilds when
          the miner switches to dev fee and back to the user job.
          Note that not all possible coins are recognised by the miner,
          and the up-to-date list is maintained on the project's webpage.

          Supported coins (the list is not exhaustive):
          `autolykos2`: erg, blocx
          `kawpow`: aipg, clore, neox, xna, rvn
          `ethash`: ethw, octa, xpb

          When dual or triple mining the value should be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

          Examples:
          --coin rvn
          --coin aipg
          --coin [1]octa

  -o, --url <URL>
          Sets the pool URL

          Format:
          <mining protocol>+<transport protocol>://<pool hostname>:<port number>

          Mining protocols:
              stratum
              ethproxy
              ethstratum
              zmp (zil only)
              solo (xelishash only)
          Transport protocols: tcp, ssl

          When dual or triple mining the value should be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.
          Failover pools are set by using `-o` multiple times.

          Examples:
          -o stratum+ssl://pool.woolypooly.com:3112
          -o zmp+tcp://us1-zil.shardpool.io:3333
              mine Zilliqa to shardpool using ZMP protocol
          -o [1]stratum+tcp://eth.f2pool.com:6688 -o [2]stratum+ssl://pool.woolypooly.com:3112
              mine the primary algorithm to f2pool and the second algorithm to woolypooly

  -u, --username <USERNAME>
          Sets the username for pool authorisation

          When dual or triple mining the value should be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

  -p, --password <PASSWORD>
          Sets the password for pool authorisation

          When dual or triple mining the value should be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

  -w, --worker <WORKER>
          Sets the worker name

          When dual or triple mining the value should be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

  -d, --devices <D1,D2,...>
          Sets GPU devices to use for mining

          Must be a comma-separated list of device indices. If not specified,
          all detected devices will be used. Devices are ordered by their
          PCI bus addresses. First GPU has index of 0.
          To display all available devices, use --list-devices.

          Example:
          -d 0,2,4,5

      --list-devices
          List available mining devices and exit

      --kernel <KERNEL1,KERNEL2,...>
          Sets CUDA kernel for selected algorithm(s)

          Some algorithms have multiple implementations with different characteristics,
          for example, one may be more energy efficient and another can achieve
          higher hashrate at a cost of increased power consumption. In such cases
          the user can choose the implementation that suits their requirements.

          Supported algorithms and values:
          nexapow
              1 - default kernel
              2 - some 20xx and 30xx GPUs may benefit from choosing this kernel
                  as more energy efficient than #1 when targeting low power draw

          sha256ton
              1 - default kernel
              2 - some 40xx GPUs may benefit from choosing this kernel
                  as more energy efficient than #1 when targeting low power draw

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

          Example:
          --kernel _,_,2,2
              GPU#0 and GPU#1 will use their respective default kernels
              GPU#1 and GPU#2 will use kernel #2

      --tune <TUNE1,TUNE2,...>
          Sets tuning values for selected algorithm(s)

          Supported algorithms:
              xelishash

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

          Example:
          --tune 184,_,320
              GPU#0 will use tuning value 184
              GPU#1 will be auto-tuned
              GPU#2 will use tuning value 320

      --dual-mode <MODE1,MODE2,...>
          Controls GPU behaviour in dual mining mode

          Format: <algo>:<dual ratio>
          `<algo>` must be one of the following:
              a1  - first algorithm
              a2  - second algorithm,
              a12 - both

          `<dual ratio>` is optional, and, if present,
          must be
          `rXX` (`dual ratio` coefficient set to `XX`), or
          `hXX` (first algorithm hashrate will drop to `XX`%)

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --dual-mode a1,_,a12:r5.2,a12:h92,a2
              GPU#0 will mine the primary algorithm
              GPU#1 will dual mine with default settings (primary algorithm at 95%)
              GPU#2 will dual mine, dual ratio set to 5.2
              GPU#3 will dual mine, primary algorithm hashrate at 92%
              GPU#4 will mine the second algorithm

      --temp-limit <LIMIT1,LIMIT2,...>
          Sets GPU thermal limits

          Sets the temperatures of GPU core and memory upon reaching which the miner
          stops the overheated GPU and waits for it to cool down before resuming.

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Examples:
          --temp-limit tc[60-70]
              stops the GPU when its core temperature reaches 70 degrees Celsius
              and resumes mining when it cools down to 60 degrees

          --temp-limit tm[105-115]
              stops the GPU when its memory temperature reaches 115 degrees Celsius
              and resumes mining when it cools down to 105 degrees

          --temp-limit tc[60-70]tm[105-115]
              enables both core and memory temperature limits

      --cpu-check
          Enables CPU verification of found shares before sending them to the pool

      --hashrate-avg <SECONDS>
          Hashrate averaging window in seconds. Default is 10.

      --power-avg <SECONDS>
          Power usage averaging window in seconds. Default is 10.

      --cclock <FREQ1,FREQ2,...>
          Sets GPU core clock frequency offset in MHz

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --mclock <FREQ1,FREQ2,...>
          Sets GPU memory clock frequency offset in MHz

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --lock-cclock <FREQ1,FREQ2,...>
          Locks GPU core clock frequency to the given value in MHz

          Comma-separated list of values can be used to set values per-GPU
          Set to `X` to reset the locked clock (unlock)
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --lock-mclock <FREQ1,FREQ2,...>
          Locks GPU memory clock frequency to the given value in MHz

          Comma-separated list of values can be used to set values per-GPU
          Set to `X` to reset the locked clock (unlock)
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --pl <PL1,PL2,...>
          Sets GPU power limit to the given value in Watts

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --mt <MT1,MT2,...>
          Sets memory tweaks for Pascal GPUs with GDDR5/GDDR5X memory

          Possible values: 1, 2, 3, 4, 5, 6 (REFRESH will be set to `mt * 16`)

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

      --fan-control <FAN1,FAN2,...>
          Sets GPU fan control mode

          Can either force the GPU fans to work at a given static speed in %,
          or adjust the fan speed to keep the GPU temperature (core, memory, or both)
          at a desired level (auto-fan mode)

          Format:
              --fan-control N
                  set fan speed to `N`%
              --fan-control t:[Tc;Tm][Finit;Fmin-Fmax]
                  enables auto-fan with `Tc` and `Tm` core and memory temperatures
                  respectively, set initial fan speed to `Finit`% and limit the
                  fan speed range to [`Fmin`%,`Fmax`%] range.
                  every value is optional and can be set to underscore `_`
                  meaning the default value should be used.
              --fan-control X
                  let the driver/OS manage the fan speed (auto)

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`.
          When dual or triple mining the value may be prepended with
          the algorithm index `[<index>]`. Primary algorithm has index 1.

          Examples:
          --fan-control 30
              sets GPU fan speed to 30%
          --fan-control t:[60;_][50;30-95]
              target core temp is 60 degrees, initial fan speed is 50%, 30%-95% range
          --fan-control t:[_;90][_;_-95]
              target memory temp is 90 degrees, max fan speed 95%
          --fan-control t:[70;100][_;_-_]
              adjust fan so that core temperature <= 70, memory temperature <= 100,
              with no restrictions on the fan speed

      --dag-reset-mclock <on/off>
          Enables or disables resetting memory overclock during DAG generation. Default is "on".

          If enabled, the miner sets the GPU memory clock offset to 0 MHz before
          building the DAG to avoid its corruption, and re-applies the original offset
          after the DAG has been built.

          Example:
          --dag-reset-mclock off
              disables memory OC reset (useful when OC is controlled by an external tool)

      --reset-oc
          Reset all overclock (including fan control) to default settings and exit

      --autolykos2-prebuild <on/off>
          Enables or disables autolykos2 dataset prebuild. Default is "on".

          If the prebuild is enabled for a GPU but there is not enough free
          memory to hold two datasets, the miner will fall back to using a single
          dataset and prebuild will be disabled.

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --autolykos2-prebuild on,off,on
              enables prebuild for GPU#0 and GPU#2 and disables for GPU#1

      --ethash-cache-dag <on/off>
          Enables or disables ethash DAG caching. Default is "on".

          If enabled, the miner attempts to keep DAGs for different epochs in
          memory to avoid rebuilding them in the future if another job with the same
          epoch is received again.
          The only limiting factor of how many DAGs can be kept is the amount of
          memory on the GPU.

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --ethash-cache-dag on,off,on
              enables DAG cache for GPU#0 and GPU#2 and disables for GPU#1

      --kawpow-cache-dag <on/off>
          Enables or disables kawpow DAG caching. Default is "on".

          If enabled, the miner attempts to keep DAGs for different epochs in
          memory to avoid rebuilding them in the future if another job with the same
          epoch is received again.
          The only limiting factor of how many DAGs can be kept is the amount of
          memory on the GPU.

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --kawpow-cache-dag on,off,on
              enables DAG cache for GPU#0 and GPU#2 and disables for GPU#1

      --nexapow-small-lut <on/off>
          Enforces using small LUT for Nexa. Default is "off".

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --nexapow-small-lut on,off,on
              enforces small LUT for GPU#0 and GPU#2

      --zil <on/off>
          Enables or disables mining Zilliqa per GPU. Default is "on".

          Useful for disabling ZIL mining for individual GPUs

          Example:
          --zil off,on,on
              disables ZIL mining for GPU#0

      --zil-cache-dag <on/off>
          Enables or disables caching Zilliqa DAG. Default is "on".

          If enabled, ZIL DAG is generated when the miner starts up and
          is kept in GPU memory at all times even if it means lower performance
          for the primary algorithm in dual/triple mining mode.
          If disabled, the DAG is built on-demand when a ZIL round starts
          and then destroyed after it's finished.

          Comma-separated list of values can be used to set values per-GPU
          To skip a GPU, set the corresponding value to underscore `_`

          Example:
          --zil-cache-dag on,off,on
              enables DAG cache for GPU#0 and GPU#2 and disables for GPU#1

      --zil-countdown
          Enables Zilliqa countdown timer

      --zil-retain-hashrate <on/off>
          Enables or disables displaying ZIL hashrate between rounds. Default is "on".

      --zil-test
          Enables Zilliqa mining test mode

          In this mode the miner will connect to a virtual Zilliqa pool which initiates
          short ZIL mining sessions more frequently than a normal pool (every 30s).
          Useful for making sure the overclock settings are stable when switching
          from the primary algorithm to `zil` and back - no need to wait for the next
          ZIL window which may start in an hour.

      --zil-test-intervals <ZIL_TEST_INTERVALS>
          Sets durations of primary algorithm and Zilliqa sessions in test mode

          Default value: `[30;30]`

          Format:
              --zil-test-intervals [D1;D2]
                  set primary algo duration to `D1` seconds
                  set ZIL duration to `D2` seconds

          Example:
          --zil-test-intervals [10;60]
              the miner will mine the primary algorithm for 10 seconds, switch to ZIL,
              mine ZIL for 60 seconds, switch back, and so on

      --enable-fork
          Enables algorithm switch for hard forks

      --activate-fork
          Activates algorithm switch for hard forks

          Can be used to simulate post-hard fork behaviour and test
          overclock settings, stratum connectivity etc. before the fork

  -l, --log-file <LOG_FILE>
          Enables logging output of the miner to the specified log file

          If the log file already exists it will be appended.
          The miner uses log rotation keeping maximum of 5 log files,
          size limit is 5MB.

          Supports the following template parameters:
              `{algo}` - algorithm name
              `{ts}` - timestamp in the format "yyyyMMdd_HHmmss"
                       note when using this option the miner will be creating
                       a new log file every time it is launched

          Example:
          --log-file rigel_{algo}_{ts}.log

          Files created:
              rigel_kawpow_20240115_123240.log
              rigel_kawpow_20240116_185402.log
              rigel_kawpow_20240116_190938.log

      --log-network
          Enables logging network traffic (useful for debugging)

      --api-bind <IP:PORT>
          Enables HTTP API and binds it to the specified socket address

          Examples:
          --api-bind 127.0.0.1:5000
          --api-bind 0.0.0.0:5000

      --proxy <PROXY>
          Sets SOCKS5 proxy for all network connections including DNS lookups

          Format:
          No auth: <host>:<port>
          Auth: <username>:<password>@<host>:<port>

          Examples:
          --proxy 127.0.0.1:8091
          --proxy admin:qwerty@myproxy.server.com:8080

      --dns-over-https
          Enables pool DNS resolution using DNS-over-HTTPS (Cloudflare)

      --no-strict-ssl
          Disables SSL/TLS certificate verification

          Useful with self-hosted mining solutions where the mining pool
          provides a self-signed certificate.
          However, adding the certificate to the system's trust store
          should be preferred (see https://github.com/rigelminer/rigel/issues/130).

      --stats-interval <SECONDS>
          GPU stats reporting interval in seconds. Default is 20.
          Set to `X` to disable periodic reports.

      --long-timestamps
          Enables milliseconds timestamps in the miner output

      --no-pool-url
          Hides pool URL from the share submissions log
          and displays the algorithm name instead

      --no-colour
          Disables colour in the miner output (TUI would still display colours)

      --no-tui
          Disables terminal user interface (TUI)

      --multi-device-arg-mode <MULTI_DEVICE_ARG_MODE>
          Defines argument validation in multi-GPU mode. Default is "match".

          Applicable to all per-GPU arguments (`--cclock`, `--mclock`, `--temp-limit`,
          `--kernel` etc.) when the number of supplied values is more than one and
          doesn't match the number of GPUs - this usually occurs when one of the GPUs
          disappears from the system (taken out for maintenance, PCIE connectivity loss
          and so on) but the miner is run with the same config.

          Accepted values:
              match - display an error and exit if the number of values doesn't match
                      the number of GPUs exactly
              first - ignore any extra values if too many, and use the first value in the list
                      for the GPUs that don't have a value
              last  - ignore any extra values if too many, and use the last value in the list
                      for the GPUs that don't have a value

          Example (assuming 3 GPUs available):
              --multi-device-arg-mode first
                  "--lock-cclock 1100,1200,1300,1400" will result in
                      GPU#0 1100
                      GPU#1 1200
                      GPU#2 1300
                  "--lock-cclock 1100,1200" will result in
                      GPU#0 1100
                      GPU#1 1200
                      GPU#2 1100

      --no-watchdog
          Disables miner watchdog

  -h, --help
          Print help information (use `-h` for a summary)

  -V, --version
          Print version information

Support

Discord: https://discord.gg/zKTgcGgc6k
BitcoinTalk: https://bitcointalk.org/index.php?topic=5424675.0