A $4,187 Bitcoin Hashboard, a $153 Monthly Bill, and the Math Sazmining Hopes You Won't Do

I opened the Sazmining product page on a Tuesday night expecting the usual. Stock photos of server racks. A hashprice calculator that defaults to the rosiest settings. An email capture box. What I found instead was one of the more candid sales pages in mining, and that's exactly what made me nervous.

Here's the offer. You buy one Proto immersion hashboard for $4,187. It hashes at roughly 204 terahashes per second at about 15 joules per terahash. Three boards make a full Proto Immersion Module, the unit Block and its partners normally move in bulk. You can start with one. That's the whole pitch, and it's a real change from the way industrial hardware has been sold to individuals.

The math nobody puts on the product page

Let's start with the power draw, because this is where the whole thing lives or dies. At 204 TH/s and 15 J/TH, the board pulls about 3,060 watts, call it 3.06 kilowatts. Run that for 30 days and you're at roughly 2,200 kilowatt hours. Divide the $153 hosting fee by those 2,200 kWh and you get about 7 cents a kilowatt hour.

That number is the entire offer. Seven cents is a respectable industrial power rate, and it tells you the hosting fee is close to a pass-through on electricity plus a thin operational margin. Sazmining isn't hiding a fat markup in the hosting line. But it also isn't giving you anything. You're renting a seat at their power contract, and the meter never stops.

Now the financing. Sazmining advertises $1,397 up front and 18 monthly payments of $155 at a stated 0% interest. Do the arithmetic and you land at $4,187, exactly the sticker price. So the 0% is real, not a fee dressed up as an APR. Credit where it's due.

But here's what the payment schedule doesn't show you. For the first 18 months you're paying $155 on the hardware plus $153 in hosting. That's $308 a month going out the door, and it's due whether Bitcoin is at $110,000 or $70,000. The financing doesn't lower the cost of mining. It lowers the cost of starting.

Then there's revenue. At a network hashrate near 1 zettahash and BTC somewhere around $90,000, a 204 TH/s board pulls in roughly $8 to $10 of Bitcoin a day. Call it $250 to $300 a month before fees. Subtract the $153 hosting charge and you're netting somewhere between $100 and $150 a month, assuming 100% uptime, which no machine on earth delivers.

Run that out and the payback on the hardware lands in the 30 to 40 month range. That's before difficulty climbs. That's before the next halving in 2028 cuts the block subsidy again. And that's before you account for a board that will be worth a fraction of $4,187 in three years, because every generation of mining silicon gets cheaper per terahash.

What Block actually gets out of this

Pull the camera back and the deal makes more sense from the other side of the table. Block built Proto, and Block needs Proto boards deployed at scale to prove the design works. Selling one hashboard at a time to retail buyers is a distribution strategy dressed up as a product launch. It widens the funnel for the hardware without Block having to build a retail sales channel of its own.

Sazmining gets something better. It gets a recurring hosting contract attached to a machine it doesn't have to buy. Every board sold is $153 a month of predictable revenue, and hosting margins compound far more reliably than hardware margins do. The story the pitch deck won't tell you is that the hardware sale is the loss leader and the power contract is the business.

And the buyer? The buyer gets optionality, which is worth something. Compare this to what retail got in 2021. Back then you handed a company $10,000, they handed you a yield number, and you never saw a machine, a serial number, or a power bill. When those outfits froze withdrawals, customers had no asset to point at. A Proto board is a physical object with a serial number. You can take delivery of it. In the event of a blowup, you're a creditor with a piece of equipment, not a name on a list.

That distinction matters more than the hashprice. It's the difference between owning a depreciating asset and owning nothing.

What I'd actually tell a friend

Two opinions, and I'll hold both.

First, the 0% financing is the least interesting number on the page. The $153 is the story. If power goes to 9 cents, or the host adds a fee, or uptime slips to 92%, your net margin gets eaten alive while the hardware payment stays fixed. Ask what the service rate actually is, ask who bears the risk if it rises, and ask for the uptime history in writing. If a salesperson can't answer those three questions in plain language, walk away.

Second, and this one will annoy some people. Most retail buyers would still be better off buying spot Bitcoin. Mining is a bet on operational execution, on cheap power, on firmware, on thermal management. When you buy a hosted board, you're betting on somebody else's execution, and you're paying them a monthly fee for the privilege. The only scenario where this beats simply stacking sats is one where you're confident in a sustained BTC price rise and you want leveraged exposure with a defined downside. That's a real strategy. It's just not the one most buyers think they're running.

So who wins here? Block wins if the boards sell. Sazmining wins if they sell and stay plugged in. The buyer wins only if Bitcoin cooperates and the power rate holds, which is a lot of ifs stacked on a $4,187 bet.

I asked myself why anyone would take this trade in a market where spot BTC is one click away. The answer is that people don't buy hashboards for the yield. They buy them because they want a piece of the machine. That's an old instinct, and it's the reason this product will find buyers regardless of what my spreadsheet says.