Skip to content

2.1.3.giga26 Gigahorse 2.0 Node / Farmer
Compare
Choose a tag to compare
@madMAx43v3r madMAx43v3r released this 22 Dec 14:39
· 64 commits to master since this release
v2.1.3.giga26
d0776ac

New in this release:

  • Chia 2.1.3 update
  • Increased compute cache size to make sure it's large enough for C19 / C20.

New since giga24:

  • Implemented recompute cache, to avoid duplicate compute when the chain re-organizes (signage point spam)
  • Both harvesters and chia_recompute_server will cache compute results, so updating just chia_recompute_server is good enough.
  • Full proof compute is now also 6 to 12% faster (ie. partials)

New since 2.1.1:

  • Bladebit compressed plot support using the GH compute engine (same performance as GH plots one level lower, BB C7 = GH C6)
  • Fixed Bladebit dropping of proofs, 3% more rewards for BB C7, 1.5% more for C6, etc.
  • Fixed Bladebit farming issue that causes ~0.5% invalid partials for BB C7, 0.25% for C6, etc.
  • Remote compute support for Bladebit plots (using the same chia_recompute_server)

New since giga14:

  • Gigahorse 2.0 plots: Compression Levels C11 to C20
  • Improved farming performance by 2x or more for C6 to C9
  • New ProofOfSpace farm benchmark to test maximum farm size for a certain partial difficulty (set via --diff)
  • CUDA disk plot mode via -3 (8G RAM enough for K32)
  • CUDA plotter Direct IO option -D for final copy (local and remote, default disabled)

Limitations:

  • The new C11 to C20 is only supported with the CUDA plotter for now
  • OpenCL farming support for the new C11 to C20 will come in a later release
  • CPU farming the new format is only supported for C11 to C13
  • The old C1 to C7 is still better for CPU farming at this point
  • Both Node & Farmer & Harvester need to be updated at the same time (when updating from before 2.1.1)

Notes:

  • Partial difficulty is now important for maximum farm size, depending on compression level.
  • Use ProofOfSpace farm --diff 50000 to benchmark for solo farming.
  • chia_recompute_server is backwards compatible to old clients, but needs to be updated for new clients running giga20.
  • Bladebit plots have the same fee as Gigahorse plots

This release provides Chia Blockchain binaries to farm compressed plots created with the new plotters provided in this repository.

The compressed plot harvester and farmer are not compatible with the official Chia node, it only works together with the Gigahorse node. However it's possible to use a wallet from the official Chia repository, instead of the Gigahorse binary wallet.

Both NFT and OG plots are supported, as well as solo and pool farming (via the official pool protocol). Regular uncompressed plots are supported as well, so you can use the Gigahorse version while re-plotting your farm.

The dev fee is as follows:

  • 3.125 % when using GPU(s) to farm compressed plots
  • 1.562 % when using CPU(s) to farm compressed plots
  • 0 % for regular uncompressed plots

The following packages are needed to run on Linux: sudo apt install ocl-icd-libopencl1

Make sure to choose the correct version for your system, see uname -m, usually it's x86_64. aarch64 is the same as arm64.

On Windows:

Note: Newer AMD GPUs starting with RX 6000 series do not work due to a driver bug, we have to wait for AMD to fix it. Vega series is known to still work when using an older driver with amdgpu-install --opencl=pal,legacy. RX 5000 series is unknown at this point.

Note: See limit-gpu-usage on how to select an OpenCL platform if you have multiple platforms (excluding Nvidia).

Assets 5