From MPW to Production: What Changes After First Silicon

For many teams, MPW feels like the goal.

First silicon arrives. The chip powers up. Basic functionality works. There is a strong sense of relief, and sometimes the assumption that the hardest part is over.

In reality, MPW success is the beginning of a different phase, not the end of the ASIC journey.

This article explains what actually changes when moving from MPW to production, why teams often underestimate this transition, and how to plan for it without unpleasant surprises.

Why MPW exists (and what it is not)

MPW exists to reduce early risk.

It allows teams to validate architecture, basic functionality, and integration at a fraction of the cost of full production masks. It is an excellent tool for learning.

What MPW is not:

• A production-quality cost model
• A guarantee of yield
• A replacement for production test planning
• A signal that the design is “done”

Treating MPW silicon as production-ready is one of the most common sources of disappointment later.

What “success” means at MPW stage

MPW success is often binary: does the chip work or not?

Production success is not binary. It is statistical.

At MPW stage, teams typically focus on:

• Power-up
• Interface functionality
• Core feature validation
• Basic performance sanity checks

At production stage, the focus shifts to:

• Yield
• Parametric margins
• Corner cases
• Manufacturability
• Test coverage and cost

These are different problems, requiring different preparation.

Yield becomes a first-class concern

MPW runs are not optimized for yield learning.

Low yield at MPW does not necessarily mean the design is bad. High yield at MPW does not guarantee production yield.

In production, yield directly affects:

• Unit cost
• Supply reliability
• Ramp schedule

Moving to production means:

• Analyzing systematic vs random defects
• Understanding layout sensitivities
• Iterating on design-for-manufacturability

Teams that ignore yield until production are forced into reactive fixes.

Test strategy changes completely

MPW test is usually minimal.

The goal is validation, not cost optimization. Patterns are limited, coverage is partial, and test time is often not representative of production reality.

Production test must balance:

• Coverage
• Test time
• Equipment cost
• Long-term repeatability

This requires:

• Early test architecture decisions
• Collaboration between design and test teams
• Iteration after silicon characterization

Production test planning that starts after MPW often becomes a schedule bottleneck.

Specification discipline becomes non-negotiable

MPW tolerates flexibility.

Production does not.

Once you move toward production masks:

• Late feature changes become expensive
• ECOs ripple through verification and signoff
• Every change carries yield and schedule risk

Teams must shift mindset from exploration to control.

This often requires:

• Freezing feature scope
• Formal change management
• Clear ownership of trade-offs

The transition is organizational as much as technical.

Cost structure shifts in non-obvious ways

Many teams expect production to simply be “MPW at scale.”

It is not.

Production introduces:

• Full mask costs
• Wafer pricing sensitivity to yield
• Packaging and test cost optimization
• Inventory and ramp planning

Some costs go down with scale. Others appear for the first time.

Understanding this shift early prevents budget shock.

Timelines stretch differently than expected

Teams often underestimate how long it takes to move from “working silicon” to “shippable product.”

Common hidden timeline drivers include:

• Silicon characterization across corners
• Yield learning cycles
• Test program refinement
• Qualification and reliability testing

Production readiness is a process, not a date.

Why teams get stuck between MPW and production

The most dangerous phase is the gap between MPW success and production commitment.

Teams hesitate because:

• Issues are visible but not catastrophic
• Budgets increase sharply
• Risk feels harder to quantify

Without a clear plan, projects stall or drift.

Successful teams treat MPW as a decision point, not a victory lap.

How experienced teams plan the transition

Teams that move smoothly from MPW to production do three things well:

• They define success criteria for MPW in advance
• They plan production test and yield learning early
• They decide, explicitly, whether and when to commit to production

The decision is conscious, not emotional.

What to do next

If you are planning an MPW run, or have just received first silicon, the most important question is not “Does it work?”

It is “Are we ready for production, and what still needs to change before committing?”

Answering that requires stepping back and re-evaluating assumptions about volume, cost, risk, and readiness.

Next step

Before moving from MPW to production, run the 2-minute ASIC or Not? Decision Wizard to confirm whether full production is the right next step now and what gaps need to be closed.

👉 /asic-or-not