NP
NurseryPool
NUMN · PoScan
← pool

Numen (NUMN) pool settings

Everything specific to this pool. Numen uses PoScan, a proof of work where each nonce grows a deterministic 3D asteroid and the proof is the spectral scan of its shape.

connection

Point your miner here:

numn.nurserypool.com:5588

Numen addresses start with nu and are 49 characters long. Create one with the official web wallet at testnet.wallet.numen-network.org , selecting Numen in the network menu at the top right, not Numen Testnet.

The pool speaks its own protocol, signed end to end. No other mining software can connect: a PoScan seal is not a block header, so nothing on the market understands it.

windows

curl -O https://numn.nurserypool.com/dl/numn-miner.exe numn-miner.exe numn.nurserypool.com:5588 YOUR_ADDRESS rig_name 8

Windows warns on any unsigned binary. Click More info, then Run anyway.

linux

wget https://numn.nurserypool.com/dl/numn-miner-linux chmod +x numn-miner-linux ./numn-miner-linux numn.nurserypool.com:5588 YOUR_ADDRESS rig_name 8

To keep it running after you close the terminal, start it inside screen or tmux.

arguments

In order: the pool, your NUMN address, a rig name of your choice, and the number of threads. Leave the last one out and the miner takes every core.

CPU only. PoScan works in double-precision floating point throughout, routed through a soft-float library so that every machine agrees bit for bit. Consumer GPUs run FP64 at a fraction of their FP32 speed, so a graphics card brings nothing here. Work is measured in scans per second: one scan is one complete asteroid, mesh generation, spectral scan and hash.

pool settings

algorithmPoScan
port5588
share difficultyvariable, about one every 8 s
pool fee1 %
miner dev fee1 %
payout threshold5 NUMN
maturityat finality, about 3 blocks

how this pool works

Share difficulty adapts to each miner so that everyone submits roughly one share every eight seconds. A share is always credited at the difficulty it was issued at, never at the one it happened to reach.

PPLNS. When the pool finds a block, it is split between the shares present in the window at that moment, in proportion to the work each represents. The window holds the most recent shares covering twice the network difficulty, with a floor of ten minutes.

Two things follow. A miner who stops before a block is found gets nothing from it, however long they mined before: their shares have left the window. And a miner who has just arrived is already earning, since their shares enter the window straight away.

Two fees, 2 % in total. One percent goes to the pool for running the node and the infrastructure. One percent is the miner dev fee, for the mining software itself, which is roughly 2.4 times faster than the reference implementation on the same hardware. Both are taken from the block before the split, and both appear separately on the dashboard.

Every share is verified. The pool recomputes the scan itself rather than trusting what the miner reports, and it does so for all of them, not a sample.

A block is only credited once GRANDPA has finalised it, about thirty seconds. Before that a block can still be reorganised out, and paying on it would mean paying on nothing. Each split is published: for every block you can see the window, the total work in it, and every miner's share.

troubleshooting

The miner connects but no job arrives. The pool stops handing out work whenever its node loses the network, rather than let you mine on a chain nobody else sees. It resumes on its own.

My address does not appear. Stats take up to a minute after your first accepted share. Check the address starts with nu and has no typo.

My hashrate looks different from my miner's. The dashboard averages over several minutes, your miner shows instant readings. Compare over an hour, not over a minute.