Miner Health Check: Is Your ASIC Performing at 100%? (2026)
Use the SoloFury dashboard Diff column to check whether your ASIC is performing at full capacity, compare identical miners, and spot a failing hashboard early.
How do you know if your ASIC is actually performing at 100%? The most honest answer is on the SoloFury dashboard, in the Diff column. Because the pool holds your share rate roughly steady, the difficulty assigned to your worker tracks your real hashrate — so a miner drifting below its peers is underperforming, even when it reports a healthy number and shows “Online.” This guide teaches you to read the Diff column, compare miners the reliable way, and spot a failing hashboard before it costs you.
Key Takeaways
- Diff is proportional to real hashrate. Vardiff holds your share rate roughly constant, so a faster miner is assigned a higher difficulty — making Diff a more honest signal than self-reported hashrate.
- Compare, don’t predict. The absolute Diff a miner settles at depends on the pool’s vardiff target; the reliable diagnostic is comparing identical miners to each other, not to a formula.
- A miner can lie with “Online.” A unit can show Online status and a plausible hashrate while a board has failed — its low Diff gives it away.
- Best Diff is luck, not health. It is a cumulative lottery statistic; a high Best Diff does not mean a better miner.
- Most low-Diff faults are thermal, power, or a dead hashboard — work the checklist in order before replacing anything.
What do the dashboard columns mean?
Enter your wallet on the SoloFury dashboard and you see a worker table:
| Column | What it means | What to look for |
|---|---|---|
| Worker | Your miner’s label | All your miners listed |
| Hashrate | Current mining speed | Within about 10% of spec |
| Best Diff | Highest-difficulty share ever found | Pure luck — not a health metric |
| Last Share | Time since the last accepted share | Seconds, not minutes |
| Shares | Cumulative shares submitted | Always increasing |
| Diff | Difficulty currently assigned to this worker | The key performance signal |
| Status | Online / Warning / Offline | Green “Online” |
What is the Diff column and why does it matter?
The Diff column shows the share difficulty the pool’s variable-difficulty (vardiff) system has assigned your worker. The insight that makes it powerful: in steady state, your Diff is proportional to your real hashrate. Vardiff adjusts each worker’s difficulty so that shares arrive at a roughly constant rate, which means a miner hashing twice as fast is handed roughly twice the difficulty. A miner that quietly loses a hashboard submits shares more slowly, vardiff lowers its difficulty, and the drop shows up in the Diff column.
How does vardiff set the Diff?
When a worker first connects, vardiff starts it at a low difficulty, then adapts: if shares arrive too fast it raises the difficulty, if too slowly it lowers it, until the share rate settles into the target band. A powerful ASIC ramps up quickly because at a low difficulty it would flood the pool with shares.
The underlying relationship is fixed by probability: a difficulty-1 share takes about 2³² (~4.3 billion) hashes on average, so the expected shares per second equal roughly hashrate ÷ (Diff × 2³²). Rearranged at a constant target share rate, Diff scales linearly with hashrate. The exact value a miner settles at depends on the pool’s vardiff target, which is why you should compare miners rather than try to predict an absolute number.
How do you estimate expected performance?
Skip absolute formulas — they depend on the pool’s internal vardiff target and mislead more than they help. Use proportionality instead. Because Diff scales with hashrate, the ratio Diff ÷ hashrate should be about the same for every identical miner on the same settings. An S21+ at 235 TH/s should stabilize at roughly twice the Diff of an S19k Pro at 120 TH/s, and far above a 1.2 TH/s Bitaxe — and two S21+ units side by side should land close to each other.
That comparative ratio is the whole diagnostic: a miner whose Diff-to-hashrate ratio is well below its identical siblings is the one to inspect.
How do you read a real fleet?
Here is an example of four identical Antminer S21+ units mining BCH, taken from a real fleet.
Example 1 — all healthy:
| Worker | Hashrate | Best Diff | Current Diff | Status |
|---|---|---|---|---|
| Miner-A | 295 TH/s | 7.593 G | 343,687 | Online |
| Miner-B | 295 TH/s | 31.108 G | 338,815 | Online |
| Miner-C | 279 TH/s | 6.582 G | 301,953 | Online |
| Miner-D | 269 TH/s | 7.031 G | 287,598 | Online |
The four settle within a roughly 17% Diff spread, which is normal for same-model ASICs. Miner-B’s Best Diff (31.108 G) towers over the others, but that is lottery noise — Best Diff is cumulative and luck-driven, not a sign Miner-B is healthier. The roughly 10% hashrate variance is also normal; judge the trend over hours, not individual minutes.
Why is comparing identical miners the best test?
When you run several identical ASICs, side-by-side comparison is the single most reliable diagnostic you have — any meaningful gap between matching hardware almost always means a fault.
Example 2 — one miner underperforming:
| Worker | Hashrate | Best Diff | Current Diff | Status |
|---|---|---|---|---|
| Miner-A | 295 TH/s | 7.593 G | 343,687 | Online |
| Miner-B | 295 TH/s | 31.108 G | 338,815 | Online |
| Miner-C | 279 TH/s | 6.582 G | 301,953 | Online |
| Miner-D | 180 TH/s | 1.205 G | 115,000 | Online |
Miner-D reports “Online,” but its Diff is nearly 3× lower than its siblings and its hashrate sits about 40% below spec. That combination is a clear hardware problem — most often a failed or throttling hashboard, a clogged heatsink, or thermal throttling.
What are the warning signs of a problem?
Watch the dashboard for these red flags:
- Diff well below similar miners (say, under half of its siblings) — a failed hashboard or thermal throttle.
- Hashrate swinging wildly (±30% within minutes) — a power-supply issue or unstable network.
- Last Share creeping past 60 seconds — the miner is stalling or disconnecting.
- Status flipping Online → Warning → Online — an intermittent connection or a pool-config error.
- Best Diff frozen at a low value for hours while shares barely move — the miner is connected but not really hashing (a firmware bug).
- Acceptance rate below 99% — high latency (wrong region), bad firmware, or hardware errors.
What do you do when performance is low?
Work this checklist in order — the cheap, common causes are first.
- Compare temperatures. Open the miner’s web interface and read chip temperatures. Anything above 85°C on stock firmware means throttling. Clean the fans and check ambient air temperature.
- Check power. Verify PSU output and that no breaker tripped. A failing PSU shows up first as periodic hashrate dips.
- Restart the miner. Often the simplest fix. Reboot from the web interface; if nothing improves in 10 minutes, power-cycle.
- Test individual hashboards. Most ASIC interfaces show per-board hashrate. On a three-board unit like the S21+, a single dead board cuts total output by about a third.
- Try a firmware change. Braiins OS+ or LuxOS often match or beat stock and expose better per-board diagnostics; switching can both help and reveal the fault.
- Check the stratum region. A miner in Europe pointed at the Americas endpoint racks up latency stale shares, which drags its Diff down deceptively. Switch to the matching regional endpoint (for example
eu-bch.solofury.com). SoloFury also offers TLS endpoints if you want the connection encrypted — see the TLS Stratum Mining guide. - Replace the unit. If board tests fail and you have ruled out power and thermals, a hashboard or control board may be defective — RMA it if it is under warranty.
For the meaning of every metric and broader troubleshooting beyond the Diff column, see the companion guide Reading Your Worker Stats.
Frequently asked questions
What does the Diff column actually tell me? It shows the share difficulty the pool assigned your worker. Because vardiff holds your share rate roughly constant, that difficulty is proportional to your real hashrate — so Diff is a reliable, hard-to-fake indicator of how much work your miner is truly doing.
Why do two identical miners show slightly different Diff values? Some variation is normal. Same-model ASICs differ a few percent in real hashrate, vardiff settles each one independently, and luck adds noise over short windows. A spread of roughly 15–20% between identical units is fine; a unit sitting far below the rest is the one to check.
One of my miners has a much lower Diff than the others — what’s wrong? A Diff well below identical siblings, especially alongside a low hashrate, usually means a failed or throttling hashboard, a clogged heatsink, or thermal throttling. Compare chip temperatures and per-board hashrate in the miner’s interface first.
Is a high Best Diff a sign of a healthy miner? No. Best Diff is the luckiest share you have ever submitted — a cumulative lottery statistic. One miner can show a far higher Best Diff purely by chance while performing identically to its peers. Use current Diff, not Best Diff, to judge health.
Why does my Diff jump around right after connecting? For the first several minutes, vardiff is still finding your level and the dashboard shows the most recent share’s difficulty, so the value swings. It should settle into a steady band within 10–30 minutes. Persistent large swings after that point are worth investigating.
My hashrate looks fine but my Diff is low — which do I trust? Trust the Diff. Hashrate on the dashboard is estimated and can be skewed by how shares arrive, while Diff comes from your actual submission rate. A normal hashrate with a low Diff often means stale shares or an intermittent fault.
What chip temperature is too high? On stock firmware, sustained chip temperatures above 85°C typically trigger throttling and lost hashrate. Improve airflow and lower ambient temperature; if one board runs much hotter than the others, suspect that board.
What’s the fastest way to spot a bad ASIC in a fleet? Sort your workers by Diff and look for an outlier. Among identical miners on the same settings, the one whose Diff (and Diff-to-hashrate ratio) sits well below the pack is your problem child — inspect it first.
Next Steps
- Open the SoloFury dashboard and compare Diff across your identical miners right now.
- Read Reading Your Worker Stats for every metric and full symptom-to-fix troubleshooting.
- If you suspect a configuration issue, revisit the Antminer S21+ Setup Guide or the Solo Start configurator.
- Use the Network Radar to confirm you are mining the most rewarding coin for your hashrate.