Ferraris and Fabrics
- 6 days ago
- 4 min read
The next wave of returns may come from the roads of computing, not just the engines.
By Oliver Campbell │ HPQC │ May 2026
“Written by humans, please don’t blame the robots for our typos”
Ferraris and Fabrics
Living in Singapore, you see a lot of sports cars. Ferraris, Lamborghinis, Maseratis, Porsches and more — they're everywhere. Last year, around1 850 of these vehicles were sold here, representing nearly two percent of all new car registrations. Step outside the supercar bracket and the volumes are larger still — BMW and Mercedes alone accounted for over ten thousand units — nearly a quarter of the new car market. These are staggering numbers in a city where the licence to own a car costs over a hundred thousand dollars, and that is before you've even considered the vehicle itself.
And yet, if you own a Ferrari in Singapore, there is a reasonable chance you will never really drive it. The roads are congested, the junctions frequent, the speed limits enforced. Whether you're sitting in a Ferrari or a Ford, you are quite likely travelling at exactly the same speed — not because the engine isn't capable, but because the infrastructure around it simply won't allow it to perform to its full potential. So it is with compute.
The data on hyperscaler capital expenditure is well known: over seven hundred billion dollars2 projected for 2026 alone, with each of the four largest technology companies now exceeding a hundred billion dollars in annual infrastructure spend. GPU3 (graphical processing unit) hardware could currently stand between 25-40%4 of data centre capex and looks to grow apace or even rise5 as AI workloads come to dominate. These are extraordinary numbers, and the impact is real — we feel it in the productivity tools we use, in the automation reshaping industries, in the pace of scientific discovery accelerating around us.
Nvidia’s Data Centre Revenue (US$ Billions)
And yet the faster we expand raw compute capacity, the more clearly the bottlenecks reveal themselves. More GPUs expose the limits of interconnects. More power demand strains grids that were never designed for this scale. More data movement collides with the fundamental physics of copper. The paradox of this moment is that the engines are extraordinary — and the roads can't keep up.
The constraints are both physical and financial, and they are tightening simultaneously. You cannot simply convert a one-gigawatt facility into a two-gigawatt one. Semiconductor node shrinkage is approaching the limits of what silicon lithography can deliver at a viable cost as Moore's Law (the number of transistors on a microchip doubles every two years) collides with Rock’s Law (the capital cost of a semiconductor fabrication plant doubles roughly every four years). And even the most elegantly designed compute architecture will underperform if the fabric connecting it is a bottleneck: copper interconnects carry inherent bandwidth and latency ceilings that no amount of GPU investment can overcome. This is where the next wave of opportunity lies — not necessarily in building bigger engines, but in building better roads.
TSMC Foundry Process Technologies
At HPQC, we look for technologies that expand what compute can actually do, not merely what it contains. Photonics, which replaces electrical signals with light to move data faster and with a fraction of the energy cost. Novel silicon architectures that do more with less, rethinking the chip not as a scaled-down version of what came before but as something designed from first principles for the workloads of today. Quantum computing, still early but advancing, is offering the prospect of solving classes of problems that classical architectures cannot approach efficiently, regardless of scale. And new interconnect fabrics — the tunnels and bridges of the compute world — that allow distributed systems to behave more like unified ones.
The hyperscalers will continue to spend. The GPU race will continue. These things matter, and we follow them closely. But we believe that some of the most durable commercial returns of the next decade will accrue not necessarily to those who build the fastest car, but to those who build the infrastructure that finally lets it drive.
Sources & Further Reading
Singapore Land Transport Authority, New Vehicle Registrations Statistics, 2026. https://www.lta.gov.sg/content/dam/ltagov/who_we_are/statistics_and_publications/statistics/pdf/MVP02-2_New_Cars_by_make.pdf
Tom’s Hardware, Google, Microsoft, Meta, and Amazon Capex Spending to Hit US$725 Billion in 2026, Up 77% from Last Year, 2026. https://www.tomshardware.com/tech-industry/big-tech/big-techs-ai-spending-plans-reach-725-billion
NVIDIA Corporation, What’s the Difference Between a CPU and a GPU?, NVIDIA Blog, originally published 2009. https://blogs.nvidia.com/blog/whats-the-difference-between-a-cpu-and-a-gpu/
NVIDIA Corporation, Fiscal Year 2026 Investor Presentation, 2026. https://s201.q4cdn.com/141608511/files/doc_financials/2026/q4/10K-NVDA.pdf
Dell’Oro Group, Data Center Capex to Grow at 21 Percent CAGR Through 2029, PR Newswire, 2025. https://www.prnewswire.com/news-releases/data-center-capex-to-grow-at-21-percent-cagr-through-2029-according-to-delloro-group-302522554.html
NVIDIA Corporation, Fiscal Year 2026 Investor Presentation, 2026. https://s201.q4cdn.com/141608511/files/doc_financials/2026/Q426/NVDA-F4Q26-Quarterly-Presentation.pdf
Taiwan Semiconductor Manufacturing Company, Foundry Technology Roadmap, 2026. https://www.tsmc.com/english/dedicatedFoundry/technology/logic
Oliver Campbell
Investment Principal, ICM HPQC Fund
May 2026
Important Note:
The information in this article should not be considered an offer or solicitation to deal in ICM HPQC Fund (Registration number T22VC0112B-SF003) (the “Sub-fund”). The information is provided on a general basis for informational purposes only and is not to be relied upon as investment, legal, tax, or other advice. It does not take into account the investment objectives, financial situation, or particular needs of any specific investor. The information presented has been obtained from sources believed to be reliable, but no representation or warranty is given or may be implied that it is accurate or complete. The Investment Manager reserves the right to amend the information contained herein at any time, without notice. Investments in the Sub-fund are subject to investment risks, including the possible loss of the principal amount invested. The value of investments and the income derived therefrom may fall or rise. Past performance is not indicative of future performance. Investors should seek relevant professional advice before making any investment decision. This document is intended solely for institutional investors and accredited investors as defined under the Securities and Futures Act (Cap. 289) of Singapore. This advertisement or publication has not been reviewed by the Monetary Authority of Singapore.
ICM HPQC Fund is a registered Sub-fund of the ICMGF VCC (the VCC), a variable capital company incorporated in the Republic of Singapore. The assets and liabilities of ICM HPQC Fund are segregated from other Sub-funds of the VCC, in accordance with Section 29 of the VCC Act.
Get to know us better:
Comments