📈 Get daily crypto insights that make you smarter about your money

The Sub-10 J/TH Epoch: Inside Bitmain’s 1,160 TH/s S23 Breakthrough and Ethereum’s 39.1 Million Staked Moat

The mining and staking landscape in late May 2026 is undergoing a profound efficiency compression. With Bitcoin firmly trading at $76,569.00, institutional operators are locked in a relentless capital expenditure race to deploy next-generation sub-10 J/TH hardware. Concurrently, Ethereum’s proof-of-stake layer has quietly swallowed 39.1 million ETH—even as the asset hovers at $2,095.10—creating a massive supply sink that is redefining the economics of blockchain security.

By Michael Nguyen | May 24, 2026

The Hardware/Software Landscape

The race for computational supremacy has officially shattered the 10 joules per terahash (J/TH) barrier, fundamentally altering the baseline requirements for industrial-scale Bitcoin mining. The catalyst for this shift is the impending rollout of Bitmain’s Antminer S23 series, slated for widespread delivery in 2026. Setting a new industry benchmark, the flagship Antminer S23 Hydro 3U delivers a staggering 1,160 TH/s while consuming just 9.5 J/TH.

This release represents the first time a single commercial mining unit has crossed the highly coveted 1 Petahash (PH/s) threshold. To put this engineering feat into perspective, the S23 series offers a roughly 40% efficiency improvement over its predecessor, the Antminer S21 XP, which was released in late 2024 operating at 13.5 J/TH. Even the air-cooled variant of the new generation, the Antminer S23 (Air), manages to hit the 9.5 J/TH efficiency mark while pushing 318 TH/s.

Software ecosystems are aggressively pivoting to manage these high-density setups. Custom firmware developers and pool operators are optimizing algorithms specifically to balance the thermal load of hydro-cooled and immersion systems. The industry is moving away from purely maximizing hashrate to dynamically under-clocking and over-clocking chips based on real-time grid pricing—a software-driven necessity for surviving the current competitive environment where margins are razor-thin.

Hashrate & Difficulty

The imminent deployment of these hyper-efficient S23 units is creating severe forward pressure on network hashrate and difficulty. As enterprise miners race to plug in machines capable of 1,160 TH/s per chassis, the broader network is bracing for consecutive upward difficulty adjustments.

While Bitcoin’s price at $76,569.00 provides a comfortable operating buffer for state-of-the-art facilities, it offers little protection for legacy operations. The introduction of 9.5 J/TH hardware means that machines from just three years ago are rapidly approaching structural obsolescence. Facilities still relying on machines operating above 25 J/TH are finding themselves increasingly uncompetitive unless they possess exceptionally cheap—or effectively free—stranded energy.

The hashrate expansion is no longer driven by merely adding more machines to a rack; it is driven by replacing entire fleets with hardware that produces exponentially more hashes per square foot. This density scaling is shifting the balance of power toward well-capitalized public miners who can secure massive hardware purchasing agreements and weather the resulting difficulty spikes.

Profitability Metrics

Profitability in 2026 is sharply divided between proof-of-work (PoW) mining margins and proof-of-stake (PoS) yields. For Bitcoin miners, the revenue per terahash is continually squeezed by rising difficulty, making access to sub-cent power agreements the ultimate competitive moat. Even at $76,569.00, the capital expenditure required to purchase S23 Hydro 3U rigs requires multi-year amortization schedules.

In stark contrast, the profitability metrics for Ethereum staking rely on capital lock-ups rather than hardware depreciation. Despite Ethereum trading at $2,095.10, the demand to generate native yield has never been higher. Validators are currently capturing rewards ranging between 2.5% and 4.2% APY, heavily dependent on base fee burn rates and maximum extractable value (MEV) optimization.

The institutional appetite for these yields is laid bare by the network’s backlog. Recent data shows a steady stream of new validators entering the activation queue. Investors are demonstrably willing to lock up capital for months just to gain entry, signaling a strong preference for predictable staking yields over the volatile margins of hardware-intensive mining operations. Other high-cap networks are seeing similar stickiness, with assets like Solana ($85.20) and BNB ($654.74) maintaining robust staking participation despite broader market fluctuations.

Environmental Impact

The environmental narratives surrounding blockchain security are diverging more sharply than ever. In the mining sector, the shift to 1,160 TH/s hydro-cooled units like the S23 series is not purely about performance; it is a thermal management mandate. Air cooling is reaching its physical limitations, prompting a massive transition toward liquid immersion and direct-to-chip hydro-cooling.

These advanced cooling systems are allowing miners to operate in hotter climates and significantly improve their power usage effectiveness (PUE). Furthermore, the ability to capture and repurpose the low-grade waste heat generated by hydro-systems for industrial and agricultural applications is becoming a standard feature of modern mining facility design, rather than an experimental fringe concept.

On the staking side, Ethereum’s environmental footprint remains negligible, a fact that is driving its massive institutional adoption. With roughly 39.1 million ETH—representing approximately 32% of the total supply—now locked in staking contracts, a massive portion of the crypto economy is secured via capital rather than electricity. Over a million active validators are processing transactions with an energy footprint comparable to standard enterprise data centers, appealing heavily to ESG-constrained funds.

Strategic Outlook

Looking ahead, the infrastructure backbone of the digital asset market is hardening into an enterprise-only arena. The millions of ETH validator queue is perhaps the most telling metric of 2026: a clear indicator that institutional capital views Ethereum (even at $2,095.10) as a foundational internet bond. With nearly a third of all ETH removed from liquid circulation, the potential for a severe supply squeeze remains high if retail demand accelerates.

Key data points shaping the immediate future:

  • Hardware Monoliths: The rollout of 9.5 J/TH equipment guarantees a new baseline for Bitcoin network difficulty.
  • Supply Sinks: With 39.1 million ETH locked and a 60-day entry queue, liquid supply on centralized exchanges is hitting historic lows.
  • Yield Disparity: The 2.5% to 4.2% APY from Ethereum staking offers a low-overhead alternative to the brutal CapEx cycles of Bitcoin mining.
  • Alternative Hubs: Capital continues to seek yield across other dominant PoS networks, keeping billions locked in assets like Solana ($85.20) and Cardano ($0.2421).

For miners, the strategic imperative is clear: secure next-generation hardware and hydro-cooling infrastructure, or be priced out of the block reward entirely. For stakers, the strategy is one of accumulation and patience, riding the yield curve while the locked supply restricts market liquidity. As both sectors mature, the days of the casual participant are decisively over, replaced by an era of industrial-scale efficiency and multi-billion-dollar staking pools.

The cryptocurrency market remains highly volatile. This article is for informational purposes only and does not constitute financial advice.

🌱 FOR BUSINESSES BitcoinsNews.com
Reach 100K+ Crypto Readers
Sponsored content, press releases, banner ads, and newsletter placements. Put your brand in front of Bitcoin's most engaged audience.

16 thoughts on “The Sub-10 J/TH Epoch: Inside Bitmain’s 1,160 TH/s S23 Breakthrough and Ethereum’s 39.1 Million Staked Moat”

  1. joule_counter

    1160 TH/s from a single machine is insane. the old S9 miners pulling 14 TH/s must feel like a bad dream now

  2. 39.1 million ETH staked at $2,095 is over $81 billion worth of ETH locked. The sell pressure absorbed by that supply sink is underappreciated.

    1. 39.1M ETH staked is a supply sink but its also a concentration risk. when the unlock wave hits after a major crash the sell pressure will be unlike anything we have seen

  3. sub-10 J/TH is the new baseline and anyone still running S19s is mining at a loss. bitmain timing this with BTC at $76k is not a coincidence

    1. S19s are already being decommissioned in texas. electricity costs make them unprofitable below $85K BTC basically. the S23 efficiency jump is existential for small miners

    2. asic_hunter S19s underwater below 85K BTC is the math nobody wants to hear. small miners selling S19s at scrap prices while industrial fleets preorder S23s by the thousand

  4. the real story here is the efficiency compression forcing small miners out. only industrial operations can afford to refresh hardware every 18 months

  5. s19_graveyard_

    S19s underwater below 85K BTC while S23 does 1160 TH/s. small miners are selling rigs at scrap prices and nobody in the article mentions what happens to network decentralization when only industrial fleets survive

    1. s19_graveyard_ exactly. 82x hashrate jump per unit in 7 years and difficulty just eats it. the small miner is dead and bitmain designed it that way

  6. 39.1M ETH staked at $2,095 is 81B locked. the concentration among 10 entities is the part nobody wants to talk about

  7. the ETH staking moat at 39.1M is impressive until you realize the validator set is concentrated among like 10 entities. decentralization theater

  8. sub 10 J/TH sounds great until you price in the hydro cooling infrastructure. the S23 Hydro needs serious facility upgrades not just a plug

  9. 39.1 million ETH staked is wild. the supply sink argument gets stronger every quarter but the price still says 2095. staking withdrawals capped the upside somehow

  10. 1160 TH/s from one machine vs the old S9 at 14 TH/s. thats an 82x jump in hashrate per unit in barely 7 years. no wonder difficulty keeps climbing

    1. validator_ops

      39.1M ETH staked at 2095 is 81B locked. the concentration among 10 entities is the real issue nobody addresses

  11. S23 Hydro at 1160 TH/s with sub 10 J/TH makes every S19 in operation a space heater burning money. industrial consolidation incoming

Leave a Comment

Your email address will not be published. Required fields are marked *

BTC$65,225.00+1.2%ETH$1,949.45+3.7%SOL$76.32+1.8%BNB$572.96+0.6%XRP$1.11+0.5%ADA$0.1654+0.1%DOGE$0.0727+0.2%DOT$0.8168-0.3%AVAX$6.70-0.8%LINK$8.76+4.1%UNI$3.90+6.4%ATOM$1.40+0.9%LTC$47.32+1.3%ARB$0.0820-0.6%NEAR$1.84+2.7%FIL$0.7515+2.1%SUI$0.7179+0.3%BTC$65,225.00+1.2%ETH$1,949.45+3.7%SOL$76.32+1.8%BNB$572.96+0.6%XRP$1.11+0.5%ADA$0.1654+0.1%DOGE$0.0727+0.2%DOT$0.8168-0.3%AVAX$6.70-0.8%LINK$8.76+4.1%UNI$3.90+6.4%ATOM$1.40+0.9%LTC$47.32+1.3%ARB$0.0820-0.6%NEAR$1.84+2.7%FIL$0.7515+2.1%SUI$0.7179+0.3%
Scroll to Top