Thor P2 review: 10 TH/s on the desk, measured at the wall
The Thor X1 amazed me with its efficiency, on a single BM1373 chip. Hammer now releases its big brother, the Thor P2: two chips, about 10 TH/s rated, a big fan and, at last, an Ethernet port. I measured it at the wall in its two most powerful profiles. The surprise comes from the normal profile.
Published 28 October 2026 · by Pixel
The measurements are my own, taken with a wall meter. The links to Helium Deploy are partner links: if you buy through them, I earn a commission at no extra cost to you.
Spec sheet
Hammer Miner2 × BM1373, the Thor X1’s chip10 TH/s · 153 W · 15.3 J/TH9.3 to 9.5 TH/s · 149 W · about 15.85 J/THabout 10 TH/s for 155 to 166 W, about 40 dBabout 35 dB in Low profile, 44 dB at 10 TH/s2.4 GHz Wi-Fi and Ethernet (RJ45)12 V 20 A (240 W) external brick included, XT60 plug, mains cord for your countryabout 15 × 18 × 15 cm$379 + $17.99 shipping to France, about €365 delivered (October 2026)- 10 TH/s measured for 153 W at the wall: 15.3 J/TH, slightly better than the 155 to 166 W sellers quote.
- The cheapest TH/s of my reviews: about €36 per TH/s delivered, roughly two to three times less than the Thor X1.
- An Ethernet port on top of Wi-Fi, which the Thor X1 lacks.
- A 240 W external power supply included, with close to 90 W of headroom and an XT60 plug built for high currents.
- Less efficient than the Thor X1: 15.3 J/TH, against 10.6 for the X1 in normal mode.
- The normal profile saves almost nothing: 149 W for 9.4 TH/s.
- According to the seller, any turbo or boost mode counts as overclocking and voids the warranty.
- Over 150 W around the clock: about €268 of electricity a year at €0.20 per kWh.
The Thor X1’s big brother
The Thor X1 was one of the first machines with the BM1373 chip, and it is still the best efficiency I have measured on a desk miner: my Thor X1 review. The Thor P2 keeps that chip but fits two of them, for about 10 TH/s: three times the X1 pushed to the max.
What else changes: a big backlit fan, an external XT60 power supply and above all an Ethernet port, which the X1 lacked.
Unboxing and power supply
The machine comes mounted on its stand, with a 240 W external power brick (12 V, 20 A) and its mains cord, matched to the delivery country. The brick connects through an XT60 plug, the yellow connector used on model cars and drones, built for high currents. That is more reassuring than a USB-C cable for a machine drawing 150 W day and night: the X1 makes do with a 65 W USB-C brick. With 153 W measured at the wall, the brick keeps close to 90 W of headroom.
Ethernet: why it matters for solo
The Thor P2 keeps 2.4 GHz Wi-Fi, handy on a shelf. But when mining solo, the machine gets new work from the pool with every new block on the network. A cable avoids Wi-Fi dropouts and losses, and so seconds spent hashing on a block that is already stale. For a machine running around the clock, it is the connection to pick whenever you can.
Getting started
On Ethernet, plug in the cable and the power: the machine gets an address from your router by itself, then find it in the router’s list of connected devices and open its interface in a browser. On Wi-Fi, the first setup works like other desk miners: give it your home network, then open its interface.
There you set the pool, with your Bitcoin address as the user, followed by a dot and a machine name, and x as the password. For solo, the FindMyBlock settings are in this guide. You also pick the power profile there.
Wall-meter measurements
Measured at the wall once the machine had settled, in its two most powerful profiles:
10 TH/s for 153 W, i.e. 15.3 J/TH9.3 to 9.5 TH/s for 149 W, i.e. 15.7 to 16 J/TH (15.85 on average)about 10 TH/s for 155 to 166 W, i.e. 15.5 to 16.6 J/THAt 10 TH/s the Thor P2 draws 153 W at the wall: slightly less than the listings say. You will also see figures around 11 J/TH: those come from a listing at 6.6 TH/s for 72 W, not from the 10 TH/s profile.
The efficiency shown by the machine itself looks better: as on the X1, it leaves out power supply losses. Only a wall meter tells you what you really pay.
The surprise: the normal profile
Look at the two profiles’ settings: 720 / 225 in normal, 728 / 226 in overclock. They are almost the same. The normal profile is already set close to the maximum, and switching to overclock costs only 4 W more (about +2.7%) for 0.5 to 0.7 TH/s more (about +6%).
The result: overclock is more efficient than the normal profile, 15.3 J/TH against 15.85 on average. My advice: leave it in overclock. And if you really want to draw less or cut the noise, switch to the Low profile (about 35 dB) or lower the frequency and voltage by hand: the normal profile saves almost nothing.
One caveat: Helium Deploy states on its listing that any turbo or boost mode counts as overclocking and voids the warranty. If the warranty matters to you, stay on the normal profile while it runs: you only give up about half a TH/s.
Versus the Thor X1
Same chip, two philosophies:
- Hashrate: 10 TH/s for the P2, 2.1 to 3.22 TH/s for the X1, three to almost five times more.
- Efficiency: the X1 is still the champion, 10.6 J/TH in normal mode, against 15.3 for the P2.
- Price per TH/s: about €36 for the P2 delivered, against €67 to €105 for the X1 (€215 to €220).
- Connection: Wi-Fi and Ethernet on the P2, Wi-Fi only on the X1.
- Noise: about 44 dB measured for the P2 at 10 TH/s (40 dB rated), against 51 dB rated for the X1 in overclock.
The full duel, figures side by side: Thor P2 vs Thor X1.
Odds of finding a block
At 10 TH/s, about 1 chance in 1,910 per year solo on Bitcoin (network as of 17 October 2026). Three to five times better than a Thor X1, but still a lottery ticket. The full math: the calculator.
On Bitcoin Cash, whose network is about 270 times smaller, the same 10 TH/s give about a 13% chance of finding a block within a year, for a block worth about $930: solo mining Bitcoin Cash.
As for electricity, 153 W around the clock is about €268 a year at €0.20 per kWh, and as much heat in the room.
The price
At Helium Deploy, the Thor P2 is listed at $379, on sale from $449. At checkout with a French address, the order added $17.99 of shipping and no tax: $396.99 in all, about €365 delivered (October 2026). It is shown in stock, shipping within 2 business days from the shop’s European Union warehouse, with a three-month warranty. It also states that VAT is not included in the listed price and that any taxes are the recipient’s responsibility: keep that in mind.
At about €36 per TH/s, it is the cheapest TH/s of my reviews.
My verdict
Get it if you want the most hashrate for your money, in a fixed machine on Ethernet that runs around the clock. And once the warranty is over, switch it to overclock: for 4 W, it is its best setting.
Go for the Thor X1 instead if every watt counts, or if you want a small, discreet machine in a corner of the desk: it is still the efficiency champion. The P2 is the champion of TH/s per euro.
A question about this review? Ask it on the Discord, several of us answer there.
Other reviews
- Thor X1 review: the first BM1373 chip at home The best efficiency I have measured on a desk miner, with closed software.
- NerdOctaxe review: 8 BM1370 chips on a desk Two NerdQAxes in one case, close to 10 TH/s, for enthusiasts.
- NerdQAxe++ Hydro review: the water-cooled mini miner Four chips, pre-assembled liquid cooling, and office-level noise.
- Bitaxe Gamma GT Turbo review: two chips, and a fan to replace Twice the hashrate of a Gamma, but a loud stock fan: a €20 Noctua fixes it.
- NerdQX 8T review: the desk miner that runs silent 7.8 TH/s for 165 W under 45 dB, a serious power supply and room to overclock.
- Bitaxe Gamma review: the open-source miner to start with About 1.2 TH/s for fifteen-odd watts, open software and a huge community.
- Avalon Nano 3S review: the Bitcoin miner that heats the room 6 TH/s rated for 140 W, quiet, and its heat works as a space heater.
- NerdNOS review: the €30 mini Bitcoin miner A NerdMiner with a real chip: 200,000 times faster, under 10 W, powered over USB-C.
Head to head: Thor P2 vs Thor X1