Configuring popular miner software.
The fastest path from "I have a GPU or ASIC" to "shares are landing" on the testnet demo. Five steps.
01 The three things every miner needs
Whatever miner you run, the pool needs the same three values: the Stratum URL of the listener, your wallet address as the username, and a placeholder password. The address is where the testnet coin units are attributed — the pool is non-custodial and never holds funds, so there is no separate login. Across every example below the pattern is identical:
The hosts shown are example placeholders such as eu.example-pool.net — substitute the hostname you stood the pool up behind. The one live demo is the GCC (Scrypt) testnet pool; the algorithm you pick must match the coin the listener serves.
02 cgminer — ASIC, SHA-256 and Scrypt
cgminer drives ASIC hardware — the SHA-256 machines used for Bitcoin-family chains and the Scrypt machines used for Litecoin-family chains, including the GCC demo. It is a command-line tool; you pass the pool as -o, the address as -u and the placeholder as -p.
For a SHA-256 ASIC, drop the --scrypt flag — cgminer defaults to SHA-256. ASICs run at a fixed, high hash rate, so there is little to tune on the miner itself: the pool assigns a starting difficulty and adjusts it automatically (vardiff) as it sees your share rate. Most ASIC owners run the vendor firmware’s web form instead of the raw binary, but the three fields are the same — URL, address-as-worker, placeholder password.
03 xmrig — CPU, RandomX
xmrig mines RandomX, the CPU-friendly algorithm. Any modern desktop or server CPU works; there is no special hardware. xmrig is usually configured through a config.json file, though the same values can be passed on the command line.
The one knob worth setting is --threads: leave one or two CPU cores free for the operating system so the machine stays responsive. RandomX is memory-hard and benefits from huge pages — xmrig prints a warning at start-up if they are not enabled, with the exact command to fix it. Leave threads unset and xmrig auto-detects a sensible count.
04 lolMiner — GPU
lolMiner is a GPU miner covering the algorithms graphics cards are good at — KAWPOW, kHeavyHash and Autolykos2 among them. You select the algorithm with -a, then pass the usual pool, address and password.
GPU tuning happens at the card, not the pool. Each algorithm has a different VRAM footprint — KAWPOW needs a few gigabytes and grows over time, kHeavyHash is light, Autolykos2 sits in between — so check your card has enough memory for the algorithm before you start. If lolMiner reports rejected shares climbing, ease off any overclock: an unstable card produces invalid shares, which the pool discards.
05 Failover and worker names
Two configuration concepts apply to every miner regardless of algorithm.
Worker names. The text after the dot in your username (...h2vp.rig-01) labels the machine. Give each rig a distinct name so the dashboard can show its shares and last-seen time separately. The address before the dot stays the same across all your machines — that is what ties their contributions together.
Failover pools. Most miners accept more than one -o / --pool entry. List a primary listener first and one or more backups after it; if the primary stops answering, the miner rolls to the next without you touching it.
Keep the same wallet address on every failover entry. That way a rollover changes only which listener is doing the work, never which address the shares are attributed to.
06 Reading the first connection
Start the miner and watch the log. Within roughly thirty seconds you should see the connection negotiated, a difficulty assigned and the first accepted share. A healthy start looks like this:
If you see authentication errors, re-check that the username is a valid address for the coin that listener serves. If shares are submitted but rejected, the most common cause is the wrong algorithm for the coin, followed by an unstable overclock.