PART ONE — THE SURGE

The Week the Network Changed

An independent look at the BrowserCoin network — the fortnight it went from CPUs to a wall of new hashpower.

The short version

The network's hashrate rose ~50× in a fortnight — from ~36,000 guesses per second (H/s) to a peak near 1,908,731 H/s — and almost all of it arrived on one day, 14 July.
It's GPUs: five addresses that appeared that day share a near-identical winning-nonce fingerprint that no established miner has.
One of them (a60f314b) now finds up to ~40% of a day's blocks, and the number of different miners winning blocks each day fell from ~100 to ~17.
One farm, or several independent operators? The public chain can't tell — they share software, not a wallet, and no money links them.

New here? Mining BrowserCoin means computers racing to find "blocks" by guessing numbers, over and over, as fast as they can. Whoever finds a block earns the coins. That's all you need to know to read this.

Hashrate is how many guesses per second the whole network makes — a direct measure of how much computing power is aimed at BrowserCoin. More hashrate = more (or faster) machines mining, and every block gets harder for any one person to find.

The network's total mining power over time. Note the scale on the left: each line up is 10× higher than the one below — so July's near-vertical jump is even bigger than it looks, from tens of thousands to nearly two million guesses per second.

It now takes just 7 addresses to find more than half of every block mined — down from 10 a few weeks ago, and as few as 3 on the busiest recent days. Fewer players, more concentrated. (Analysts call this the "Nakamoto number.")

PART TWO — WHO ARRIVED

Who Arrived, and When

The newcomers did not trickle in. For most of the network's life a few dozen steady miners found the blocks; then, on 14 July, a cluster of addresses that had never found a single block began finding them in bulk. The lines below are each address's running total — flat until mid-July, then rising almost vertically. Organic growth does not look like this: word of mouth arrives scattered across days and weeks, not all at once at full output. One of these addresses, a60f314b, went from nothing to finding two of every five blocks mined on 16 July. And as the newcomers piled in, the number of different addresses winning any block on a given day fell from around a hundred to seventeen — the same blocks, found by far fewer hands.

Cumulative blocks by address

Each line is one mining address's running total of blocks found. Organic growth climbs gradually; these arrived at full scale within a day.

Top addresses since 11 July

post-Jul-11 blocks; addresses are real — verify any at tabscope.netlify.app

Addr	First seen	Blocks	Share	Ramp

Distinct daily winners

From about a hundred distinct winners a day down to seventeen — fewer distinct winners means the blocks are concentrating in fewer hands.

What this means if you mine here

A small CPU miner now shares the network with GPU rigs that find blocks far faster, so a solo block is harder to come by than it was a month ago. What a block pays hasn't changed — only the odds of finding one alone. Pooling smooths that out: many small miners combine hashrate and split what's found, so payouts arrive steadily instead of as a rare jackpot.

PART THREE — ARE THEY CONNECTED

Are They Connected?

Arriving on the same day is suggestive, but it is not proof of a single owner. The public chain records two things we can check on every miner: how it searches for a block, and where its coins go. How it searches is the most revealing, so we start there.

The fingerprint: how they search

To find a block, a computer guesses numbers one after another — starting low and counting up — until it hits a winning one. How high it usually has to count before it wins tells you how fast it is.

SHARE OF WINS AT THE VERY BOTTOM

A slow computer's winning guesses land all over the place. A graphics card (GPU), guessing billions of times a second, almost always finds a winner while it's still near the very bottom of the range. The long blue bars are five addresses that almost always win at the bottom — no ordinary miner does. It's like five people who all sign their name with the same rare flourish: it doesn't prove they're the same person, but it proves they're using the same unusual tool.

Money flow: the surge addresses vs. everyday pool payouts

A shared tool is not a shared bank account — so we follow the money. If these five were one operation cashing out, their coins would eventually pool together.

The five surge addresses

No transfers between them, and none to a shared wallet — there is nothing on-chain to trace.

Where the money actually moves — ordinary pool payouts.

These two addresses are pools — they fan payouts out to dozens and hundreds of miners (a few example amounts shown). That is all the money that moves; the five surge addresses above sit untouched.

The surge is unmistakably GPU hardware — the nonce fingerprint proves it, and it arrived on a single day.

Whether it is one farm or several independent operators who all adopted the same GPU miner that week, the public chain cannot tell us: they share software, not a wallet, and leave no money trail linking them. Consistent with a coordinated arrival; on-chain, indistinguishable from several independent newcomers.

How we did this — check us

We pulled all 30,600 blocks from the public helper API, decoded each one, and computed: hashrate — worked out from how hard the network had set the puzzle and how fast blocks were coming in; each address's first block and ramp; the winning-nonce distribution; and every coin transfer between the top addresses. Every number here is recomputable — the chain is public. Verify any address at tabscope.netlify.app.