65nm vs 40nm vs 28nm: How Teams Choose a Node in Practice

Choosing an ASIC node is one of the easiest ways to add cost and risk without realizing it.

Many teams assume that the most advanced node automatically delivers the best product. In practice, experienced teams do the opposite: they choose the least aggressive node that still meets the product requirements.

This article explains how teams actually choose between 65nm, 40nm, and 28nm in practice, and why node selection is a business decision as much as a technical one.

Start with the real drivers, not the node name

Node selection rarely starts with geometry. It starts with constraints.

Before looking at node options, teams that make good decisions can answer these questions clearly:

• What is the dominant driver: power, cost, performance, or longevity?
• Is this a digital-heavy design or mixed-signal?
• Which interfaces are mandatory?
• How stable is the specification today?
• How sensitive is the business case to NRE and schedule risk?

If these questions are fuzzy, any node choice is premature.

Why teams choose 65nm

65nm remains one of the most widely used nodes for a reason.

It offers mature manufacturing, stable IP ecosystems, predictable costs, and strong long-term availability. For many mixed-signal and industrial products, it is not a compromise, it is the optimal choice.

Teams often choose 65nm when:

• Power targets are moderate and manageable
• Mixed-signal content is significant
• Long product lifetime is required
• Risk and predictability matter more than density

In many cases, 65nm provides more than enough performance with far fewer surprises.

Why teams choose 40nm

40nm is often the “middle ground” node.

It delivers measurable improvements in power efficiency and density compared to 65nm, while avoiding some of the complexity and cost associated with more aggressive nodes.

Teams gravitate toward 40nm when:

• Power efficiency matters but is not extreme
• Die size impacts unit cost
• IP availability is solid for required interfaces
• They want a balance between improvement and risk

40nm is frequently chosen for second-generation products where the architecture is already proven.

Why teams choose 28nm

28nm is typically selected when power and integration density are central to the product’s value proposition.

It enables lower power per operation and higher integration, but it also demands more from the team in terms of verification, signoff, and integration discipline.

Teams choose 28nm when:

• Power efficiency is a differentiator, not just a target
• High integration reduces BOM or form factor significantly
• Proven IP exists for all critical interfaces
• The team or partner has experience at this node

Choosing 28nm without a clear driver often increases risk without delivering proportional benefit.

The hidden cost of choosing too advanced a node

A common mistake is selecting a node that exceeds the product’s real needs.

The hidden costs include:

• Higher NRE
• Longer verification and signoff cycles
• Increased sensitivity to late changes
• Fewer fallback options if assumptions change

These costs do not appear immediately, but they surface as schedule slip, budget pressure, and reduced flexibility.

A practical checklist before locking a node

Before committing to 65nm, 40nm, or 28nm, teams should be able to answer:

• Which constraint truly drives the decision?
• What would break if we chose a less aggressive node?
• Is the IP we need proven on this node?
• How stable is the spec today?
• What is our tolerance for NRE and schedule risk?

If these answers are unclear, the safest move is to pause the node decision.

The rule experienced teams follow

For a first ASIC, experienced teams usually choose the lowest-risk node that still meets the product targets.

Moving to a more aggressive node should be a deliberate choice tied to clear business value, not fear of being “left behind.”

What to do next

Node selection only makes sense once you are confident that ASIC itself is the right direction.

Before locking into a node discussion, it helps to step back and answer a simpler question: does ASIC make sense for your product now, later, or not at all?

Next step

Run the 2-minute ASIC or Not? Decision Wizard to get a directional recommendation and understand whether node selection is even the right conversation to have today.

👉 /asic-or-not