Seven contenders, one unique angle

Crystal-EM Hybrid is not the only post-Moore proposal. Seven other approaches have research momentum, capital, and roadmaps. To honestly defend the thesis we have to know what they are, where they shine, and where they break.

Key Concept

Technology Readiness Level (TRL)

A 1–10 NASA/DoD scale: TRL 1 = basic principles observed; TRL 5 = lab prototype; TRL 8 = system qualified through test; TRL 10 = mission-proven. Useful single number for 'how far is this from production?'

Key Concept

Post-Moore Computing

The umbrella term for any approach that aims to keep computing performance scaling after the end of classical CMOS feature-size reduction. Includes new materials, new architectures, new physics, and new packaging.

Diagram · Technology showdown

interactive

Tech A

Tech B

Metric	Silicon	Crystal-EM
Performance	8/10	9/10
Power Eff.	5/10	9/10
TRL	10/10	4/10
Scaling	3/10	9/10
Manufacturing	10/10	4/10
Cost	9/10	5/10

Pick any two technologies; compare across six axes.

The seven competitors, in one paragraph each

1. GAA Nanosheets (TRL 8). Industry's official next step after FinFET. Stacked nanosheets fully wrapped by gate. Strength: drop-in replacement for the existing fab. Weakness: still rides the size curve — eventual atomic floor.

2. Carbon Nanotubes (TRL 4). Single-walled CNTs as the channel. Strength: mobility ~1000× silicon. Weakness: no one can place trillions of perfectly-aligned tubes at scale.

3. 2D Materials (TRL 3). Single-atom-thick MoS₂, graphene, BlackP channels. Strength: atomically thin → ultimate gate control. Weakness: contact resistance and wafer-scale uniformity remain unsolved.

4. Photonic Computing (TRL 5). Compute with photons, not electrons. Strength: zero resistive loss, parallelism. Weakness: photons don't store state — needs hybrid electro-optical packaging.

5. Quantum Computing (TRL 3). Qubits exploit superposition. Strength: exponential speed-up for specific algorithms. Weakness: error correction overhead is enormous; not a general-purpose CPU replacement this decade.

6. Neuromorphic Computing (TRL 5). Spiking neural hardware, in-memory compute. Strength: ~100× efficiency on inference. Weakness: programming model is alien; software ecosystem is a decade behind.

7. 3D Chip Stacking (TRL 8). Stack many silicon dies vertically. Strength: shipping today (HBM, V-Cache). Weakness: heat removal — power density goes up linearly with stack height.

Where Crystal-EM Hybrid sits (TRL 4 today). Of the eight approaches above, seven still depend either on shrinking features (GAA, CNT, 2D) or on packaging tricks (3D stacking) or on radically alien programming models (quantum, neuromorphic). Crystal-EM is the only candidate whose scaling mechanism — improving crystal purity and EM coupling efficiency — has no miniaturisation dependency at all. That is the unique angle.

Checkpoint · +5 XP

Which technology shares with Crystal-EM the property that performance can rise without shrinking the device?

Same road, different cars

GAA, CNTs, 2D materials, and 3D stacking are all faster cars on the same shrinking-road. Quantum and neuromorphic are different vehicles for different races. Crystal-EM is a *new road surface* — every car already on it gets faster without changing the car.

Try It Yourself

Find the Crystal-EM crossover year

Open the Scaling Laws module. Toggle every technology curve on simultaneously. Find the year on the x-axis where the Crystal-EM curve crosses above the silicon-CMOS curve. Note: that crossover year shifts dramatically with ECCF.

Open Scaling Laws

You've completed Track 4 — The Thesis. You can now: state Simon's Law and explain its three pillars, name the experimental work (Wang, Tokyo) that grounds pillars 1 and 2, write down ECCF and explain each sub-variable, and place Crystal-EM Hybrid in the post-Moore landscape. With this, Track 5 (Hands-On Lab) is unlocked — time to put the thesis to work in the simulator.

Lesson Summary

Seven other post-Moore technologies compete for the post-silicon throne.
GAA, CNTs, 2D materials, photonics, quantum, neuromorphic, 3D stacking — each has strengths and a fundamental limitation.
Crystal-EM Hybrid is unique in one respect: its scaling mechanism does not require miniaturisation.
TRL (Technology Readiness Level) and scaling-path independence are the two axes that matter most for long-term betting.

Test Your Knowledge · +60 XP

TRL stands for:

Which post-Moore approach has the highest TRL today?

The fundamental weakness of 3D chip stacking is:

Quantum computing is best described as:

What makes Crystal-EM Hybrid uniquely positioned vs the other seven approaches?