@bitcoin_artisan First overclock hits different 😄 Just keep an eye on the chip temps — if you're creeping past ~65C, drop the core voltage a notch. Big temp drop, barely any hashrate lost.
auto-discovery re-scans on a timer, so when you power a new axe on mid-session it just appears in the list, no restart, no manual ip. small thing, but it's the difference between managing a fleet and babysitting it.
how many axes are you running right now, and at what point did you stop checking each one's web ui by hand. trying to find where single-miner habits break for people.
if you run solo, set your stratum difficulty low enough that you actually see shares land. you won't find a block faster, but a dashboard that never updates for hours tells you nothing about whether the miner's even healthy.
every reading baller takes lands in a csv. temp, voltage, hashrate, error rate, timestamped. when someone asks why your gamma fell off a cliff at 2am, you don't guess, you open the log and read it back. data beats vibes.
psu sizing for a gamma: the chip pulls maybe 18-20w, but cheap bricks droop hard at the rail under sustained draw. a 5v supply with real current headroom holds voltage steadier than a 'rated' one that's lying to you.
the stock gamma fan moves enough air until it doesn't. once ambient hits summer numbers, a 40mm fan at 100% is the gap between 1150mhz stable and a thermal throttle loop. airflow is the real overclock, not voltage.
what's the first number you check when you open your mining dashboard in the morning? hashrate, temps, shares, or something weirder? genuinely curious what people anchor on.
suggested difficulty vs a hard-set value: on a gamma, letting the pool auto-tune difficulty usually beats pinning it unless you know your share rate cold. a wrong fixed diff just floods or starves your share submissions.
@NordicMiningSwe Before buying a new cooler, the quick heatwave fix is dropping frequency/voltage a notch til it passes: small hashrate hit, big temp drop. A cheap box fan across the GT801 buys a few degrees too. Auto temp tuning for hot spells is what the whole scene's been hacking together.
quiet milestone: more of you run baller across 3+ miners than single units now. it started as a one-axe convenience and became a fleet tool because that's how you actually run them. noted, building for it.
a single gamma running baller solved a real block recently. one asic against the whole network, full reward, no pool. the odds say it shouldn't happen. it keeps happening anyway. that's the appeal.
@YuurinB@derekmross@VectorPrivacy Nice — getting all your Gammas into one view is the whole game once you're past a couple units. That local-network dashboard is exactly why I built Baller: hashrate, temps and uptime for every miner in one place. A /commands bot is a slick DIY route too though.
the thing you can't see with one axe: drift. across a fleet, two identical gammas on identical settings pull different watts and run different temps. silicon lottery is real and only obvious side by side.
the csv export is underrated. an afternoon of per-minute hashrate logs shows you the exact temperature where your unit starts shedding shares. you can't eyeball that from a live gauge.
undervolting is where the J/TH wins actually live, not overclocking. hold your frequency, walk voltage down 5mv at a time until shares drop, then back off one step. same hashrate, less wall draw, cooler vrm. free efficiency.
thermals before frequency. if your fan curve is reactive instead of predictive, every freq bump just trades hashrate for throttling. hold the die under 60c first, then push clocks into the headroom you made.
@VoskCoin You should try our app https://t.co/zTo53iXFmu on your Umbrel, we published it in out Community Umbrel store, its free and would love any feedback.
axeos and a fleet dashboard aren't competitors. axeos is the cockpit for one miner: deep, live, per-chip. the dashboard is air traffic control for all of them. two tools, two jobs. you want both.