This week, IBM (IBM  ) announced that it was able to develop the world's first 2-nanometer chip. This is considered a breakthrough in integrated circuit design as the previous smallest chip was a 3nm chip by Taiwan Semiconductor (TSM  ).

IBM's 2nm chip will be able to produce CPUs that are capable of 45% higher performance or 75% less energy use than the standard designs that are used today. Currently, Intel's (INTC  ) desktop processors are still being built on top of a 14nm process and are attempting to migrate to a 10nm chip. In contrast, competing chipmakers are already using a 7nm or 5nm standard.

Originally, process size was a good way to estimate a chip's power, because it correlates to the two-dimensional size of transistors etched onto the chip surface. However, now that chips have three-dimensional designs, it's no longer an effective indicator of a chip's power.

A more accurate measure is to look at a chip's transistor density. In essence, this is how many transistors are packed onto the chip. And, by this measure, the IBM chip really shines as it packs 333 million transistors onto 1 square millimeter.

In contrast, Intel's desktop chip has 45 million transistors per square mm, while its laptop chip has 100 million transistors per square mm. AMD's (AMD  ) leading chip has 91 million transistors per square mm, while Apple's (AAPL  ) M1 chip has 171 million transistors per square mm.

IBM believes that its new chip design could lead to cell phone battery life quadrupling. However, skeptics note that IBM's chip is months away from being produced, while TSMC's 5nm process is already in production, and 3nm chips are set to begin production next year. And, both of these chips have comparable performance measures.

Of course, IBM lacks the production capacity and would have to partner with a chip manufacturer. Reportedly, the company is in talks with Intel and Samsung although no agreement has been struck yet.

Stock Price Outlook

While the last decade has been tremendous for tech stocks, IBM has been an exception. The stock has underperformed and is down 30% from its 2013 peak.

Notably, the company's revenue peaked in 2013 at around $105 billion and has declined to $70 billion in the last quarter on a trailing 12-month basis. EPS has dropped even more from $15 per share to $6 per share.

The company has consistently tried to innovate in enterprise software, cloud computing, AI, and now chip design. However, it's failed to have a meaningful breakthrough. Until this change, IBM seems more like a value trap that investors should avoid.