Token computeInterstice Advisory
Constraint Builder
NeoCloud 50 MW — Oregon (BPA)
stableRegime II
Reference scenario — Clone to save changes
Why it bindsThe machine

Global AI demand backlog (SemiAnalysis Oct 2024: 3–5× current supply) fills site to 93% utilization. PPA at $28/MWh is below $30/MWh threshold — cost advantage sustains global customer acquisition. 12% pricing premium reflects GPU-scarce market clearing (CoreWeave H1 2025).

Marginal
NeoCloud 50 MW — Oregon (BPA)US-OR
Capital

Marginal: capital availability is binding — programme scale or returns at risk.

Supply exceeds demand by 8.7×40% premium on binding capital

Binding constraint

Financing cost or availability caps the investable programme

The project cannot proceed at target scale or requires equity substitution that materially reduces returns.

The facility costs $40B over 15 years — even empty. A 10% IRR requires $4B/year in net revenue. Very few entities can credibly underwrite that.

Near-binding risks

Hardware

30 percentage points below the binding threshold

Timeline

35 percentage points below the binding threshold

Regime 2 — Marginal: Mild supply-demand imbalance. Utilisation pressure is building but the facility remains cash-generative; one or two constraint releases would restore balance.

Regime 2 — Marginal. Binding: Capital. Near-binding: Hardware.
0.0
PowerElectrical power availability and grid capacity.
0.1
ThermalHeat rejection capacity of the cooling system.
0.0
NetworkExternal bandwidth and path-diversity constraint.
0.5
HardwareGPU and accelerator supply constraint.
0.5
TimelineConstruction and commissioning schedule constraint.
0.0
LaborSkilled workforce availability for operations.
0.2
PermitsRegulatory and permitting approval constraint.
0.8
CapitalCommitted funding availability for the next expansion.
derivedgap metric & fill per node — engine-derived from registry assumptions