One miner needs no software at all. You find its IP, open the web interface, paste in a pool, and forget about it. Our setup guide covers that in full.
Five miners is a different job. Twenty is a different job again. At that point you stop configuring machines and start managing a fleet, and the difference between doing that well and badly is measured in the hours you spend and the downtime you do not notice.
This page covers the tools that actually exist for that — the manufacturers' own utilities, third-party firmware, and monitoring platforms — with what each one really costs and what it genuinely does. Checked 12 September 2026.
Manufacturer tools: free, and the first thing to install
Every major manufacturer ships a free utility for finding and batch-configuring its own machines. These are unglamorous and they are what you should reach for first.
| Manufacturer | Tool | What it does |
|---|---|---|
| Bitmain | IPReporter | Finds a miner's IP. Run as Administrator, click Start, then hold the IP Report button on the control panel for ~5 seconds until it beeps. |
| Bitmain | APTool | Batch operations across many machines at once — scan the subnet, set pools, upgrade firmware, reboot, export a fleet list. |
| MicroBT | WhatsMinerTool | IP Monitor discovery plus batch config. Long-press the IP Found button until two LEDs flash. |
| ElphaPex | ElphaPex Tool | The equivalent utility for the DG Scrypt range — discovery, batch pool config and firmware. |
| IceRiver | IceRiver Tool | Discovery and batch configuration for the KS and AL series. |
Two things about these that catch people out.
Default credentials must be changed, and on MicroBT it is not optional. Antminers ship with root as both username and password; WhatsMiners use admin. On MicroBT the API will not function at all until you change the default password — if you are wondering why your monitoring tool cannot see the machine, that is usually why.
Batch tools operate on whatever they find. APTool and its equivalents scan a subnet and will happily apply a pool change or a firmware flash to every machine they discover, including ones you did not mean to touch. Put miners on their own VLAN or subnet before you run anything in batch mode. This is the single most common way people take down their own fleet.
Third-party firmware: real gains, real costs
Custom firmware replaces the manufacturer's software to tune each chip individually rather than running one profile across all of them. The main options are Braiins OS+, LuxOS (from Luxor) and Vnish.
The claimed benefit is meaningful — Braiins states an S21 Pro moves from roughly 15.0 to 13.2 J/TH, about a 12% efficiency gain. We have not independently verified that figure and you should treat vendor efficiency claims with the same scepticism you would apply to a manufacturer's.
The costs, by contrast, are concrete and worth stating plainly:
- A development fee of roughly 2 to 2.8% of your mined output, taken continuously. On a machine grossing $10 a day that is about $100 a year.
- Warranty. Bitmain's warranty is void if you change function settings without consent. On a machine inside its 365-day warranty, that is a bad trade.
- Risk. A failed flash can brick a control board. Practise on one machine before touching a fleet.
The honest arithmetic: a 12% efficiency gain minus a 2.5% revenue fee is worth having if electricity is a large share of your costs. At hosted rates around 7 cents it is a clear positive. At the very cheap rates our hosting facilities run, power is a smaller share of revenue and the gain shrinks toward the fee. Work it out for your own rate before flashing anything.
Our rule of thumb: in-warranty machine, leave it alone. Out of warranty, expensive power, more than a handful of machines — worth doing.
Monitoring: the thing that actually pays for itself
Efficiency tuning gets the attention. Monitoring saves more money, because the expensive failure is not a slightly inefficient miner — it is a miner that has been offline for nine days and nobody noticed.
| Platform | Model | Suits |
|---|---|---|
| Awesome Miner | Free tier, then paid by machine count | Mixed fleets, Windows-centric operations |
| Foreman | Per-miner monthly subscription | Larger deployments, strong alerting |
| minerstat | Free tier, then per-worker | Mixed ASIC and GPU |
| Braiins Farm Proxy | Free with Braiins OS+ | Fleets already on Braiins firmware |
What to configure on day one, in priority order: an alert for a machine going offline, an alert for hashrate dropping more than about 10% below nominal, and a temperature alert. Those three catch almost everything that matters. Pool-side hashrate graphs are a useful cross-check — if your monitoring says a machine is fine but the pool shows no shares, believe the pool.
If you host with us, monitoring is our job rather than yours, and uptime is covered by the terms in our SLA guide. This section is for people running machines themselves.
Error codes worth recognising
Bitmain's codes are the ones most people meet. Two patterns save real money.
| Code | Means | Usual remedy |
|---|---|---|
| V:1 – V:4 | Power initialisation, calibration or voltage error | Check the PSU before anything else |
| J1:1 – J8:1 | Bad ASIC on a chain | Replace the PSU first — see below |
| P:1 | High-temperature protection | Airflow, ambient temperature, dust |
| F-codes | Fan fault | Fan replacement or cable |
| N:4 | Network loss | Cable, switch, DHCP |
The J-code point is the valuable one. A J-code reads like a dead hashboard, and people write off boards on the strength of it. But a failing power supply presents as both V-codes and J-codes, and Bitmain's own documented remedy for J-codes is to replace the power supply. A PSU is a fraction of the cost of a hashboard. Swap it before you condemn anything.
The second money-saver: dust. Most P-codes and a good share of fan faults are dust in heatsinks and fan bearings. Compressed air on a schedule prevents more failures than any software will. Our guide on how long ASIC miners last covers maintenance intervals.
What we would actually run
One to three machines: the manufacturer tool for discovery and configuration, and nothing else. Check the pool dashboard once a week. Anything more is overhead you do not need.
Four to twenty machines: manufacturer batch tool plus a monitoring platform with offline and hashrate alerts. This is the point where monitoring starts paying for itself — one unnoticed nine-day outage on a machine grossing $10 a day costs more than a year of most subscriptions.
Twenty or more: proper monitoring with alerting to your phone, a separate VLAN for miners, a documented IP scheme, and spare PSUs on the shelf. At this scale custom firmware is usually worth evaluating too — but test it on one machine for a fortnight before rolling it out.
Whatever the count: change the default passwords, keep miners off your main network, and write down which machine is on which IP. Those three cost nothing and prevent most of the problems people actually have.
If the management overhead is the reason you are hesitating about mining at all, that is a fair reason — and it is the main practical argument for hosting rather than running machines yourself. Our home versus hosted comparison weighs both, and the catalogue lists what we stock.
Sources: Bitmain and MicroBT service documentation for tool behaviour, default credentials and error codes; Braiins published efficiency claims and development fee schedule; vendor pricing pages for monitoring platforms. Checked September 2026. Third-party firmware efficiency claims are vendor figures and are not independently verified here. Modifying firmware may void your manufacturer warranty.