High-performance CUDA GPU miner for HTN (Hoohash V110).
Supports NVIDIA GPUs (CUDA).
Dynamically linked. Built on Ubuntu 20.04.
Linux only binary, but runs fine under WSL.
Requires CUDA 12 Latest Version 1.2.6
Recommended for all modern NVIDIA cards.
Optimized x64 CPU miner for HTN (Hoohash V110).
Requires AVX2. Fully statically linked.
Linux only binary, but runs fine under WSL.
Fastest CPU miner currently available for x64.
Latest Version 1.2.5
Optimized aarch64 CPU miner for HTN (Hoohash V110).
Apple Silicon (M1 or newer).
Only CPU miner currently available for Apple Silicon.
Optimized aarch64 CPU miner for HTN (Hoohash V110).
Linux on aarch64 (ARMv8). Statically linked.
Only CPU miner currently available for ARM64.
1. Download the appropriate miner for your hardware.
2. Extract the archive:
tar -xzf hoo_gpu.tar.gz
or
tar -xzf hoo_cpu.tar.gz
3. Run the miner with your pool and wallet info:
./hoo_gpu/hoo_gpu -o stratum+tcp://your.pool:port -u hoosat:yourwallet -p x --same-stratum [-O stratum+tcp://backup.pool:port]
./hoo_cpu/hoo_cpu -o stratum+tcp://your.pool:port -u hoosat:yourwallet -p x --same-stratum [-O stratum+tcp://backup.pool:port]
-t [threads] (CPU miner) Number of CPU threads to use.
--list-gpu (GPU miner) Show available GPU devices and their IDs.
--gpu-id 0,1,2 (GPU miner) Select specific GPU devices.
--slow-cpu (GPU miner) Reduce CPU load at cost of slightly fewer blue blocks.
--same-stratum Use the user-supplied pool for DevFee instead of discovery.
--timeout [seconds] How long in seconds to wait for a stratum connection to respond (default 15 seconds).
--help List all command-line options.
Ask in the HTN Discord. Community members and devs monitor it regularly.
Yes via WSL (Windows Subsystem for Linux). A native Windows build is not planned currently. Install WSL, a supported Linux distribution, then run the miner as on Linux.
Yes, both hoo_cpu and hoo_gpu work under HiveOS and MMPOS. You can integrate them via a custom miner package, using the versioned download links above
On MMPOS you want to be runnning ./hoo_gpu rather than ./hoo_gpu/hoo_gpu
Use hoominer for AMD or Intel GPU support. This page focuses on NVIDIA CUDA GPUs and CPU mining.
Yes. All miners here have a 1.5% DevFee mined to: hoosat:qq2g85qr...
Lock memory to 810 MHz (Turing or later). Hashrate scales with core clock; 1710–1910 MHz is sensible.
Best: non FP64-crippled accelerators (P100, V100, A100, H100). Titan V is only prosumer card worth it; K80 is too old.
A higher DevFee applies to Data Centre class cards.
Core count & clock speed matter most. L3 cache size is irrelevant.
Pre AVX2 CPUs unsupported (hoo_cpu will not start).
L1 cache DOES make quite a difference, 32KB of DATA L1 cache is the absolute
minimum for decent hashrate (64Kb L1 total cache). More is better
hoo_cpu_arm can run in Termux (64‑bit).
But you almost ceratinly want our native Android app HooDroid
No. Metal does not expose usable FP64 for GPU cores; only CPU mining.
Not directly. Use the HTN Stratum Bridge.
stratum+tcp and stratum+tcps supported.
See the HTN Wiki guide.
HTN is a 5bps network. That means 200ms to find a block. Latency matters. Solo mining is the norm.
You almost certainly want to add --same-stratum to the arguments if mining to a local bridge
Sure. hoo_xpu -c config.json
An example config file:
{
"url": "stratum+tcp://192.168.1.14:5555",
"user": "hoosat:qrm2jaklpf95t4a3kxm04zr27sayq9avvdwqcapm5c696qa3vj3lj3ye3nr9s.5950X",
"pass": "x",
"same-stratum": true,
"threads": 31,
"backup-stratum": "stratum+tcps://htn-lon.foztor.net:5554"
}
-O/--backup-stratum--old-autotuner-T, --timeout=N)--cpu-threads-ignore N mine on total threads-N CPU cores--stratum stratum+tcp://192.168.1.14:5555,stratum+tcps://htn-lon.foztor.net:5554
Or --stratum "stratum+tcp://192.168.1.14:5555 stratum+tcps://htn-lon.foztor.net:5554"fatbins folder present from the tarball--stratum as alias of -o for easier hoominer compatibility--user as alias of -u for easier hoominer compatibility--cpu-threads as alias of -t for easier hoominer compatibility--rpc-url 192.168.1.14:5556 This needs to point to the HTTP RPC endpoint on your stratum bridgehealth_check_port: :5556 in config.yaml. Other ports work fine too--exp-threshold (default 1.6) (hoo_cpu ignores it currently). Range 0.1 - 2.0--allow-double-solve when pool mining --allow-double-solve--stale-limit N (Higher N allows for more staleness, but more likelihood of a block being red.) --stale-limit --no-cpu-validate is specified--slow-cpu now actually works (use when on an old CPU with many GPUs, and CPU load is problematic, will reduce hashrate slightly)--timeout
-O, --backup-stratum URL flag--same-stratum
--same-stratum flag