Quantum Computers vs. P-Computers: Cost and Implementation Speed Determine the Winner
Are you familiar with the "P-computer"?
Currently, the interest of many investors and engineers is focused on quantum computers.
Indeed, quantum technology holds the potential to fundamentally overturn the concepts of computation.
However, it is also true that high hurdles remain regarding the "physical constraints" and "manufacturing costs" required for practical application.
Now, the "P-computer" (probabilistic computer), which attempts to solve the challenges faced by quantum computers by extending existing semiconductor manufacturing technology, is quietly and surely moving into the practical application phase.
Many may believe that quantum computers are the best "next-generation computing paradigm".
However, in terms of cost, implementation speed, and compatibility with existing AI infrastructure, the P-computer has the potential to become a powerful "natural enemy" that surpasses quantum.
Can the P-computer be a rival to the quantum computer?
And what kind of "victory or defeat" awaits in the future of computing infrastructure?
From the perspective of an "investor who is an IT engineer by trade," I will unravel the structural advantages of the P-computer.
Chapter 1: What is a "P-computer"?
The "P" in "P-computer" stands for "Probabilistic," meaning it is a "probabilistic computer."
To understand the P-computer (probabilistic computer), let's first simply organize the differences between it, existing computers, and quantum computers.
The computers we use daily perform calculations by clearly deciding between "black or white," or 0 and 1.
On the other hand, quantum computers use a special physical phenomenon called "quantum mechanics" to handle both 0 and 1 states simultaneously, attempting to solve complex puzzles all at once.
So, what about the P-computer?
In a word, it is a "computer that intentionally uses thermal 'fluctuations' and 'ambiguity' for calculation."
It is not about seeking a perfect answer, but rather about being adept at finding the "statistically most likely solution" at tremendous speed.
To use an analogy, while a quantum computer "derives the shortest route using the special abilities of quantum mechanics," a P-computer "instantly picks out the best answer from countless options, armed with intuition and probability."
Both share common ground in that "probability" is the foundation of their thinking and the target they are trying to realize.
However, from an investor's perspective, there is a decisive difference between the two.
Quantum computer: Since the protagonist is "quantum," it requires extremely advanced "special materials" and a "dedicated manufacturing environment." It can be compared to the production of a custom-made, ultra-luxury car.
P-computer: It can utilize the existing manufacturing lines for "silicon semiconductors (CMOS)" used in familiar smartphones and PCs as they are. It can be compared to a hybrid car that can be mass-produced in current automobile factories.
In other words, the P-computer has an overwhelming economic advantage in that "there is no need to build a new factory from scratch."
While quantum computers chase "future dreams," P-computers can be called a new computing engine that is realistically implementable on top of "existing industrial infrastructure."
Chapter 2: The P-computer leaps over the hurdles of quantum computers
Because of their revolutionary computing power, quantum computers face "physical and economic hurdles" that must be overcome.
The first is the problem of "cooling".
Currently mainstream superconducting quantum computers require huge and expensive refrigerators to cool them down to near absolute zero.
This equipment not only incurs maintenance costs but also has limited installation locations and environments, making it a major barrier to widespread social adoption.
The second is the problem of the "manufacturing environment".
Chips for running quantum computers often require special equipment and materials that differ from the standard manufacturing processes that the existing semiconductor industry has spent decades building.
This makes it difficult to utilize the existing massive semiconductor ecosystem as is, and it is difficult to rapidly advance cost reductions through mass production.
On the other hand, the P-computer (probabilistic computer) leaps over these two hurdles by "repurposing existing technology."
The core circuit of a P-computer (the p-bit) can be manufactured using the same mechanisms as the CMOS semiconductors that power modern smartphones and PCs.
In other words, semiconductor factories around the world can be converted into 'P-computer production hubs' just as they are.
Is it a technology that assumes a 'special environment,' or one that leverages 'existing infrastructure'?
From an investor's perspective, it should be clear which one will win in terms of the speed of recouping upfront investment and the speed of social adoption.
Chapter 3: The P-Computer Breakthrough
In the previous chapter, we explained that P-computers leap over the hurdles faced by quantum computers, but P-computers also had their own unique challenges.
For many years, this concept has been held captive by the curse of physical wiring.
The manual method of connecting probabilistically fluctuating spin elements and control circuits with individual cables required an extremely delicate construction process and remained at the proof-of-concept stage of at most 100 p-bits.
However, in June 2026, a joint US-Japan research team consisting of Tohoku University and the National Institute of Standards and Technology (NIST) overcame this physical constraint.
They succeeded in fully integrating probabilistic bits (p-bits) onto a silicon substrate for the first time in the world, using existing semiconductor integrated circuit manufacturing processes.
For details, please refer to this article (Xeno Spectrum).
I believe the significance of this achievement lies in the scalability to a '1 million p-bit' level.
At the previous 100 p-bit scale, the range of calculations was likely limited to simple combinatorial problems.
However, if integration enables operations at the 1 million p-bit scale, we can see the potential to derive solutions with efficiency that overwhelms existing computers for tasks that consume vast computational resources in the real world, such as logistics optimization, molecular structure exploration in drug discovery, or complex portfolio optimization in finance.
That said, this technology will not immediately become a universal tool.
While the increase from 100 p-bits to 1 million p-bits is certainly a significant evolution, it is considered that this number of p-bits is still far from practical application, and adaptation on the software (algorithm) side that leverages the characteristics of this hardware will also be required.
However, the fact that we have been freed from the constraints of physical wiring and have obtained a mass-producible platform indicates that this is a historic breakthrough where the P-computer has graduated from the laboratory verification stage and entered a new phase of market implementation.
Chapter 4: The Impact of P-Computers on the Stock Market
The breakthrough mentioned in the previous chapter means that 'the P-computer has been connected from the sanctuary of the laboratory to the massive supply chain of the semiconductor industry.'
In this chapter, we will examine what kind of impact this fact will have on the stock market.
4.1. The Birth of a New Market Segment: 'Coprocessors'
What investors should first focus on is the point that P-computers will not 'completely replace' conventional deterministic computers (CPUs/GPUs).
Currently, for AI inference and complex optimization problems, conventional digital chips consume enormous amounts of power and time to perform calculations through brute force.
In this domain, P-computers will likely emerge in the market as 'specialized coprocessors' that derive 'probabilistic solutions close to the optimum' at high speed with overwhelming low power consumption.
What is important here is that the birth of this coprocessor has the potential to push the massive 'upfront investment project' known as quantum computing into a crisis of survival.
Until now, quantum computers have been expected to be the 'last resort' for optimization problems and complex simulations.
As of June 2026, another quantum boom is occurring, coupled with government funding from the Trump administration, and many quantum-related stocks are soaring.
However, they are still struggling with the high-cost hurdles of 'cooling' and 'manufacturing environments' mentioned in the previous chapter, and they have neither been put into practical use nor generated profits.
What will happen when a P-computer appears in the near future that can be mass-produced in existing semiconductor factories and solve optimization problems with practical energy efficiency?
For many companies and national projects, quantum computers will be downgraded to a 'dream for the distant future,' and in the real business world, the scenario where affordable and efficient 'P-computer coprocessors' take market share will become realistic.
Investment money for quantum computers will lose patience with the slow pace of practical application and flow out all at once to P-computers, which offer faster investment recovery.
This 'shift in expectations' is likely the biggest impact that could occur in the stock market over the next few years.
While diversified tech companies like Google and IBM might be able to absorb this as a localized issue, quantum-specialized companies may face a fatal crisis if they do not take action early.
4.2. 'Redefining' the Semiconductor Ecosystem
The transition of P-computers from research-level theory to the implementation phase of 'integration with CMOS manufacturing processes' will begin to bring about a decisive change in the power map of the semiconductor industry.
First, the superiority of contract manufacturing will increase further.
Until now, building a P-computer was a 'high-barrier technology' that required both advanced physics and specialized circuit design.
However, once CMOS-integrated technology like this is established, mass production will be possible with existing semiconductor manufacturing equipment.
In other words, giant foundries like Taiwan Semiconductor Manufacturing Company (TSMC), Samsung Electronics, and Intel will be able to undertake manufacturing simply by incorporating a new module called a 'probabilistic calculation block' into their existing production lines.
This 'versatility' will be the driving force for recovering massive capital investment.
As a next scenario, a new division of roles for fabless (design-specialized) companies will emerge.
The possibility of emerging semiconductor startups with proprietary algorithms and circuit design know-how specialized for probabilistic calculation rising to prominence will also increase.
This is because they do not need to own factories themselves and can use the foundry's manufacturing process as a 'canvas' to introduce P-computer-specialized chips to the market.
What investors should pay attention to here is the question, 'Who will standardize the 'probabilistic calculation block'?'.
If a company emerges that holds the circuit design IP (intellectual property) for p-bit calculation as an exclusive or de facto standard, like ARM in the CPU market, that company could build a high-profit structure that threatens existing giant tech companies.
The semiconductor ecosystem will be forced into a structural evolution from the traditional singular path of pursuing 'high-precision, deterministic computing' to a hybrid manufacturing line where certainty and probabilistic computing coexist. Companies that can respond rapidly to this shift, and those that hold next-generation design IP, will be the winners in the next-generation tech market.
The foundries and companies holding next-generation design IP that can respond quickly to this transition will be the winners of the next-generation tech market.
4.3. The 'Foundation' as an Investment Target
As of June 2026, with no pure-play P-computer stocks in the market, the strategy investors should take is to invest early in the 'peripheral infrastructure' that will inevitably support the implementation of this technology.
Let's consider which sectors will benefit the most as P-computers enter the mass production phase.
While I believe AI-related fields will undoubtedly benefit, I will exclude them from this selection on the premise that they will continue to grow even without P-computers.
The first option that comes to mind is semiconductor manufacturing equipment manufacturers. Tokyo Electron (TEL) and Applied Materials (AMAT) would fall into this category.
Integrating P-bits into CMOS processes requires special process technology to precisely place and stack MTJ (Magnetic Tunnel Junction) elements within miniaturized circuits.
Companies that upgrade existing lithography and deposition equipment and dominate the cutting-edge manufacturing processes that determine the yield of probabilistic computing chips should see their revenue opportunities explode as P-computers become widespread.
The second option is materials and components manufacturers included in the supply chain of major foundries. Examples include photo-process material manufacturers like JSR and Tokyo Ohka Kogyo (TOK), or top companies in silicon wafers and related materials like Shin-Etsu Chemical and SUMCO.
The reliability of probabilistic computing chips depends heavily on the purity and characteristics of the materials used.
Companies that are deeply embedded in the roadmaps of TSMC and Intel and can provide the special materials and photomasks required for next-generation chip structures will be valued as 'indispensable partners' in technological innovation.
The third option is hyperscalers (large-scale cloud operators).
NVIDIA, Microsoft, or Google (Alphabet) are examples.
They operate their own data centers, and reducing power consumption is a top priority.
If P-computers are implemented as coprocessors, they can dramatically improve the energy efficiency of their own servers, so the move to incorporate P-computer technology into their own chip development ahead of other companies will accelerate.
They are both customers and, in effect, the biggest patrons of technological development.
The final option is to 'decide to wait and see'.
Precisely because we are at a stage where it is impossible to know how many years away practical application and monetization are, we should not rely on speculative investments, but rather focus on gathering information.
Investors should be keenly aware of the ripple effect when 'P-computer-related news' emerges, specifically which companies' supply chains it will impact.
The true beneficiaries of the P-computer era are companies that have cleared the physical constraints of 'can they actually deliver manufacturing equipment' and 'can they provide the materials essential for manufacturing next-generation circuits'.
Chapter 5: Conclusion — The 'Investment Compass' for the P-Computer Era
As of June 2026, this technology is still in its infancy.
However, with laboratory theories breaking through the technical barrier of 'full integration onto silicon substrates,' the tide that investors should pay attention to has clearly changed.
It is premature to conclude that P-computers will replace quantum computers, but it is certainly an extremely practical and disruptive turning point that drags optimization calculations, which were bound by the 'constraints of physical laws,' down into real semiconductor manufacturing lines.
Investors need to set aside the excessive expectations for the 'dream of quantum computing' that has been talked about for years and identify the signs that investment capital is shifting toward P-computers, which have a more solid roadmap for practical application.
While quantum computers remain in the research phase as an 'ultimate dream,' the era in which P-computers begin to erode the existing semiconductor market as a practical weapon called a 'coprocessor' may arrive sooner than expected.
If you cannot wait for companies with P-computer-related stocks to go public, there is the option of early investment.
The important thing is the players who have the ability to 'actually give shape' to that technology.
Supporting extreme micro-fabrication manufacturing equipment manufacturers
Constructing next-generation circuits advanced materials manufacturers
Pursuing power efficiency to the limit hyperscalers
Lastly, technological evolution is always discontinuous and faces unpredictable phases.
That is why I recommend that you do not rush into a decision right now, but rather continue to gather information to catch reliable signals while incorporating a wait-and-see strategy.
The day when P-computers solve our social issues and make vast computational resources more efficient is approaching moment by moment.
Whether your portfolio ends up as a mere 'investment in a trend' or bears fruit as 'foresight that anticipated market changes' depends on the perspective from which you continue to observe the market starting today.