The Decentralized Compute Forum

Where the decentralized compute debate happens.

Honest numbers, strong opinions, no paywall. GPU markets, DePIN economics and the price of AI — dissected weekly by the team behind Omniverse Compute.

Hot Takes

Short, sharp, arguable

Deep Dives

Long-form breakdowns
Deep Dive The bubble question is the wrong question

Bulls and bears both assume today's GPUs stop being useful. Depreciation and obsolescence are two clocks running at wildly different speeds — and only one of them can be a bubble.

Deep Dive The 12x water multiplier: 92% of a data center's water footprint sits at the power plant

US data centers consumed 66 billion liters of water on site in 2023 and about 800 billion litres generating their electricity — a 12× multiplier that almost never appears in a sustainability deck. Once you price that, "closed loop or evaporative" stops being a values question and becomes arithmetic with a break-even you can calculate.

Deep Dive From wafers to panels: the glass substrate pivot, and who ships first

Intel's 2023 test chip claims 10x interconnect density; TSMC's June 2026 supply-chain plan tests 0.8mm glass cores at 85x110mm; Absolics builds on a NIST-backed $75M award. The next packaging wall may be glass — a stage-by-stage read of who has published what.

Deep Dive The agent token bill: what a 15x task actually charges

An agent's bill has four segments: fresh input, cached input at ~0.1x, visible output, and reasoning tokens billed as output but never shown. Anthropic measured 4x/15x multipliers; our arithmetic says raw token counts overstate the input bill by ~5x when caching works.

Deep Dive HBM ate the DRAM supply: the memory bill comes for everyone (2026)

Server DDR5 contract prices up 93–98% in one quarter of 2026, spot DDR5 die up 307% in four months, and Micron walked away from its own 29-year-old consumer brand. The AI memory shortage is displacement, not addition — and everyone downstream pays.

Deep Dive A megawatt at 800 volts: the arithmetic that retires the 54 V busbar (2026)

1 MW at 54 V is 18,519 A and 332 kg of copper per metre; at 800 V it is 1,250 A and 22 kg. But the busbar only burns 0.06% of a gigawatt — so the 3–5% efficiency claims are about deleted conversion stages, and Google and NVIDIA quote different numbers because one is a retrofit.

Deep Dive The biggest battery in America is hiding in a data center (2026)

720 Megapacks, 2.8–3.3 GWh, behind the meter at one AI campus — against a 25 MWh ride-through for a 100 MW hall. The battery stopped being a UPS and became a load-shifting generator, and an instrument for buying grid access.

Deep Dive Modular AI data centers compress 18 months into 100 days. The calendar still belongs to the grid (2026)

Alibaba Cloud's CUBE 5.0 delivers a large AI data center in 100 days (30 factory + 50 install + 20 commissioning) against a 12–18-month US norm — but the window starts at a powered shell and ends before IT arrives, while grid interconnection averages 4+ years. Modularization compresses the build, not the queue; our arithmetic says the speed benefit ≈ the cost benefit.

Deep Dive The same rack of GPUs has three list prices: $10.50, $10.58 and $27.04 per GPU-hour (2026)

One GB200 NVL72 rack — 72 Blackwell GPUs in a 130 TB/s NVLink domain — rents at three official prices: CoreWeave $10.50, AWS $10.582, Azure on-demand $27.04 per GPU-hour. Azure's own 3-year commitment collapses the anchor to $11.90; the ~$10.4M per-rack-year spread is one megawatt of construction. List is an anchor; commitments are the market.

Deep Dive The power bill has a second page: $333.44 per MW-day, an 85% floor, and a $750M deposit (2026)

PJM's 2027/28 capacity auction cleared at the FERC cap of $333.44/MW-day — 11.5× the 2024/25 price — and still came up 6,623 MW short. Capacity bills the megawatt you reserve, not the MWh you burn: $15.44/MWh at 90% load factor, $46.31 at 30%. Ohio's 85% floor for 12 years and Virginia's $1.5M/MW collateral turn that line into a balance-sheet commitment.

Deep Dive The gas turbine is sold out to 2032. Data centers are about 15% of the buyers (2026)

GE Vernova ended Q3 2026 with 122 GW of gas equipment backlog plus slot reservations — 6.1 years of output at its 20 GW run rate, which lands in 2032, the delivery year it sold slots for. Siemens Energy closed fiscal 2026 at 104 GW, 39 of them reservations. Data centers are 10–20% of buyers and 100% of the marginal price: the scarce product is a date, not a machine.

Deep Dive The AI bottleneck moved to a laser material: NVIDIA paid $4 billion for indium phosphide capacity (2026)

NVIDIA's 2026-03-02 investment of $2 billion each in Lumentum and Coherent equals roughly 4.4× Lumentum's entire fiscal-2026 capex of $451.3 million — and Lumentum's revenue ran $480.7M → $1.01B → $1.26B in twelve months while net PP&E grew 60%. The constraint on 800G/1.6T optics is wafer time in compound-semiconductor fabs, new capacity lands after the 1.6T ramp, and China refines ~70% of the world's indium. The $4B is option premium on queue position.

Deep Dive AI servers are a financing business: three builders, one cash problem (2026)

Dell's AI server revenue doubled to $16.4B with a 20.9% gross margin — and free cash flow fell 47%. Supermicro shipped $19.9B at a 12.7% gross margin and used $434M of operating cash, with receivables up $3.56B and inventory up $2.48B. The margin is rent on allocation; the constraint is working capital, and Dell's $95B backlog is 5.8 quarters of demand priced before the memory bill arrives.

Deep Dive Memory is being sold forward: 26 take-or-pay deals and a $150 billion floor (2026)

Micron ended fiscal 2026 with 26 strategic customer agreements — take-or-pay deals covering 35%+ of revenue through 2030 — and ~$150 billion of RPO counted at minimum prices, backed by $32 billion of customer deposits ($12.7 billion on its balance sheet). SK hynix signed 14 LTAs covering 50%+ of 2027–2028 capacity with demand at 2× supply. The spot price is becoming the residual, and the floor is struck at record-cycle margins.

Deep Dive Half of GE Vernova's backlog isn't machines — it's decades of labor (2026)

The $197B backlog splits into $104B equipment (+92% YoY) and $93B services (+14%) — 47% of the order book is annuity, not turbines. Services backlog grows twice as fast as services revenue converts, customer down payments drove $5.3B of the quarter's $5.4B free cash flow, and every one of the 32 turbines shipped is a future service contract being commissioned.

Deep Dive The fuel cell is sold like a server, not a power plant (2026)

Bloom Energy's Q3 2026 was $1,235.8M of revenue — 86.9% product, 5.5% service, at a 33.8% gross margin. GE Vernova books 47% of its backlog as a services annuity; Bloom books a machine. What buyers want is a date: 1 GW at $40B over four years is $27.4M a day of asset-time against a ~$280M-a-year power premium.

Pricing Watch

Real GPU market rates, tracked

Landscape

Who's who, reviewed honestly
Landscape You can't police a GPU you can't see

Export law governs chips crossing borders, not compute accessed remotely. That mismatch split the market into compliant capacity and shadow capacity — and both are now priced in.

On-Chain Gaming

Games, economies — and their compute bills
Playable Real or AI — can you tell which photo is real?

Five pairs a day. One photograph is real, one was generated. We built it because half the arguments in this forum come down to "looks real to me" — now there is a scoreboard.

Homelab

Your own machine as the datacenter
Homelab Cloud GPU vs home GPU: the 2026 TCO spreadsheet

A used RTX 3090 is about $820. Whether it beats renting depends on three numbers: your utilization, your electricity price and your exit. The full arithmetic, line by line.

Homelab Quantization is a VRAM problem, not a disk problem

The GGUF file on your disk is roughly two-thirds of what actually runs. Why the KV cache — not the weights — decides which model fits your card, with the arithmetic for 14B, 32B and 70B.

Homelab 2.5GbE vs 10GbE: what the standards actually say

2.5G was standardised in 2016 to run over the Cat5e you already own. 10G is where cable becomes the spec — and where the drive, not the network, becomes your ceiling.

Homelab VA vs watts: sizing a home-lab UPS that actually holds (2026)

A UPS is sold in VA but your rack draws watts — and APC's own docs put the industry rule at "approximately 60% of the VA rating." The two-limit arithmetic, the runtime curve, and the checklist for sizing backup power honestly.

Homelab Your SSD's TBW is a workload assumption, not a lifespan (2026)

Micron rates the same 30.72 TB drive at 56,000 TB or 14,000 TB depending on the write pattern — a 4× spread on one datasheet. The DWPD formula, why the client class assumes 8 hours a day at 40°C, and the one number to measure before buying used flash.

Homelab A home lab runs on 1,920 watts, not 2,400: the NEC 80% rule (2026)

NEC 210.20(A) sizes the breaker at the noncontinuous load plus 125% of the continuous load — and a 24/7 rack is a continuous load by definition. So a 20 A / 120 V circuit is 1,920 W continuous, not 2,400 W: the load tally, the receptacle math, the wire gauge, and the 240 V option that doubles your watts on the same copper.

Homelab A home rack's cooling problem is the room, not the fans (2026)

1,000 W of compute is 3,412 Btu/h of heat and needs about 316 CFM at a 10 °F rise — roughly six 120 mm fans at free air. Fans only set the temperature difference, and the fan laws charge 8× the power for double the airflow. The equation, the ASHRAE envelope, and why 30 m³ of room air warms a kelvin every 36 seconds.

Homelab The per-outlet limit is the connector, not the breaker: C13/C19 in a home rack (2026)

A rack PDU has three amp limits and most people check one. The branch allows 1,920 W continuous, but the C13 connector is rated 10 A by IEC 60320 (12 A on Vertiv's own UL tables) — about 1,440 W per outlet at 120 V — and C19 steps to 16 A. The failure mode is a warm plug and intermittent reboots months later, never a breaker trip.

Homelab Your PSU closes the UPS's gap: hold-up time decides the topology, not the transfer spec (2026)

A line-interactive UPS pauses its output for 2–4 ms on transfer; the Intel ATX12V spec requires a power supply to hold regulated output for 17 ms at full load, and ATX 3.1 still mandates 12 ms. The bridge your PSU was born with covers the UPS's gap three to six times over — which is why most home racks need AVR and correct sizing, not double conversion.

Homelab Your GPU server is limited by lanes, not slots (2026)

AM5 exposes 24 usable PCIe 5.0 lanes, Xeon 6 gives 96, EPYC 9005 gives 128 per socket, Threadripper PRO with WRX90 gives 144. Four GPUs at x16 need 64 of them before storage. And x8 versus x16 barely moves tokens per second — lanes buy load time and headroom, not throughput.

Guides

Hands-on: get set up, step by step

About this forum

The Decentralized Compute Forum (DCF) is an editorial publication about GPU markets, DePIN networks and the economics of AI compute. We write it, we argue in it, we publish our sources — and we change our minds in public when the data does.

No paywall, no sponsored posts. Comments run on GitHub Discussions — github.com/CBTdao.

Full disclosure: DCF is operated by the Omniverse Compute (OMC) team — a decentralized GPU network project on BNB Chain, currently in public testnet. Coverage of OMC and its competitors follows the same rules as everything else here: dated numbers, primary sources, public corrections. The privacy policy, terms and about page carry the details.