Foundry model. is contract semiconductor manufacturing in which a supplier fabricates wafers from customer designs using qualified process platforms and design rules. The pure-play form avoids selling competing end products, supporting confidentiality and trust across many fabless customers. TSMC’s creation in 1987 under Morris Chang is widely associated with establishing the scalable pure-play model that enabled design companies to form without funding their own factories. Semiconductor economics couple very large fixed commitments to uncertain product demand. Architecture, software, verification, masks, process qualification, factories, equipment, substrates, packaging capacity, test time, and inventory must be funded before lifetime volume is known. At the leading edge, design and mask nonrecurring expense can reach hundreds of millions of dollars, while a greenfield logic fab can require well above ten billion dollars and years to ramp. Mature nodes remain economically important because analog, RF, power, embedded memory, display, sensor, connectivity, and control functions do not automatically benefit from maximum transistor density. Revenue therefore depends on product mix, wafer starts, die area, yield, package complexity, utilization, pricing, customer concentration, and the timing of replacement cycles—not merely nominal node.

Business model, market position, and economics. Foundry revenue is driven by wafer starts, product and node mix, wafer pricing, utilization, yield arrangements, packaging services, and long-term agreements. A shared process spreads fab, equipment, enablement, and yield-learning cost across customers, while each mask set and design remains private. Leading-edge capacity commands high investment and often higher wafer prices; mature capacity can generate attractive returns through high utilization, depreciated assets, embedded features, long lifecycles, and disciplined expansion. Competitive advantage accumulates across reusable IP, talent, design methodology, process recipes, yield history, packaging know-how, developer tools, customer relationships, standards, and installed software. These assets reinforce one another but also create switching costs and concentration risk. A strong product can still lose if its toolchain is difficult, supply is constrained, total system cost is poor, or customers cannot qualify it in time. Conversely, an older node or architecture can remain attractive when it is stable, available, inexpensive, security-qualified, and supported for a decade. Roadmaps should be read as directional commitments; production readiness requires design kits, working silicon, repeatable yield, capacity, packaging, and customer shipments.

Technology, product architecture, and implementation. A foundry product is much more than a transistor. It includes PDK models, design rules, reference flows, standard cells, SRAM compilers, I/O, analog and interface IP, reliability models, DFM, mask infrastructure, process control, wafer sort support, and increasingly advanced packaging. Readiness progresses from research through risk production, qualification, yield ramp, volume, automotive variants, and long-term support. A node can be available while a required memory, voltage option, package, or IP block is not. A credible comparison starts at the workload and system boundary. Peak arithmetic, core count, transistor count, or process label alone says little about useful performance. Engineers examine sustained throughput, tail latency, memory capacity and bandwidth, cache behavior, interconnect topology, I/O, precision support, compiler maturity, power envelopes, cooling, reliability, security, serviceability, and software portability. For process and manufacturing choices they add density by circuit type, voltage range, SRAM scaling, analog behavior, design rules, IP readiness, yield learning, reticle limits, packaging, and qualification. Published specifications are usually conditional on product configuration and workload, so normalized measurements and clear test conditions matter.

Execution, supply chain, and engineering risk. Customers evaluate confidentiality, neutrality, geographic footprint, capacity, cycle time, defect density, parametric yield, excursion response, quality systems, packaging, technical support, and financial durability. TSMC has the largest pure-play scale; Samsung Foundry combines logic manufacturing with a broader electronics group; GlobalFoundries, UMC, and others emphasize differentiated or mature platforms; SMIC is important within China under equipment and export constraints. Market-share numbers vary by source, period, currency, and inclusion rules. The operating system behind a shipped chip spans architecture, RTL, verification, physical design, signoff, tapeout, mask preparation, wafer fabrication, probe, assembly, final test, firmware, drivers, libraries, system validation, and field support. A schedule slip in one layer can idle investment elsewhere. Capacity reservations, long-lead equipment, substrate allocation, export controls, geographic concentration, single-source materials, and qualified second sources shape resilience. Quality systems must connect inline process data to wafer sort, package test, board behavior, and field returns. Change control is especially strict for automotive, industrial, medical, aerospace, infrastructure, and other products with long service lives.

<svg viewBox="0 0 760 470" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system,Segoe UI,Roboto,sans-serif">
  <rect width="760" height="470" fill="#0d1117"/>
  <defs>
    <marker id="fmBlueArrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="6" markerHeight="6" orient="auto">
      <path d="M0 0L10 5L0 10Z" fill="#60a5fa"/>
    </marker>
    <marker id="fmGreenArrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="6" markerHeight="6" orient="auto">
      <path d="M0 0L10 5L0 10Z" fill="#34d399"/>
    </marker>
    <marker id="fmGoldArrow" viewBox="0 0 10 10" refX="8" refY="5" markerWidth="6" markerHeight="6" orient="auto">
      <path d="M0 0L10 5L0 10Z" fill="#f59e0b"/>
    </marker>
    <radialGradient id="fmWafer">
      <stop offset="0" stop-color="#c4b5fd"/>
      <stop offset=".55" stop-color="#60a5fa"/>
      <stop offset="1" stop-color="#172554"/>
    </radialGradient>
    <filter id="fmFabGlow" x="-30%" y="-30%" width="160%" height="160%">
      <feGaussianBlur stdDeviation="5" result="blur"/>
      <feMerge><feMergeNode in="blur"/><feMergeNode in="SourceGraphic"/></feMerge>
    </filter>
  </defs>

  <text x="380" y="31" fill="#e6edf3" font-size="21" font-weight="700" text-anchor="middle">The Foundry Model — Many Chip Designs, One Shared Fab</text>
  <text x="380" y="53" fill="#8b98a5" font-size="12.5" text-anchor="middle">fabless companies own products; the foundry concentrates process IP, equipment, manufacturing scale, and yield learning</text>

  <!-- Multiple product companies feed one manufacturing platform -->
  <g transform="translate(24 86)">
    <text x="103" y="-8" fill="#93c5fd" font-size="10.5" font-weight="700" text-anchor="middle">FABLESS CUSTOMERS</text>
    <rect width="206" height="239" rx="12" fill="#0b1726" stroke="#60a5fa" stroke-width="1.6"/>

    <!-- AI chip floorplan -->
    <g transform="translate(14 18)">
      <rect width="178" height="54" rx="7" fill="#142b44" stroke="#60a5fa"/>
      <rect x="9" y="8" width="39" height="25" fill="#1d4f72" stroke="#38bdf8"/>
      <g fill="#1f3d5a" stroke="#60a5fa">
        <rect x="53" y="8" width="25" height="12"/><rect x="82" y="8" width="25" height="12"/><rect x="111" y="8" width="25" height="12"/>
        <rect x="53" y="24" width="25" height="12"/><rect x="82" y="24" width="25" height="12"/><rect x="111" y="24" width="25" height="12"/>
      </g>
      <text x="157" y="19" fill="#93c5fd" font-size="8.5" font-weight="700" text-anchor="middle">AI</text>
      <text x="157" y="32" fill="#8b98a5" font-size="7.2" text-anchor="middle">accelerator</text>
      <text x="89" y="48" fill="#8b98a5" font-size="7.2" text-anchor="middle">architecture · RTL · product software</text>
    </g>

    <!-- MCU floorplan -->
    <g transform="translate(14 85)">
      <rect width="178" height="54" rx="7" fill="#10251f" stroke="#34d399"/>
      <rect x="9" y="8" width="41" height="29" fill="#15382d" stroke="#34d399"/>
      <rect x="55" y="8" width="75" height="13" fill="#17382d" stroke="#34d399"/>
      <rect x="55" y="25" width="34" height="12" fill="#17382d" stroke="#34d399"/>
      <rect x="93" y="25" width="37" height="12" fill="#17382d" stroke="#34d399"/>
      <text x="155" y="19" fill="#6ee7b7" font-size="8.5" font-weight="700" text-anchor="middle">MCU</text>
      <text x="155" y="32" fill="#8b98a5" font-size="7.2" text-anchor="middle">controller</text>
      <text x="89" y="48" fill="#8b98a5" font-size="7.2" text-anchor="middle">low power · embedded flash · long life</text>
    </g>

    <!-- RF floorplan -->
    <g transform="translate(14 152)">
      <rect width="178" height="54" rx="7" fill="#211936" stroke="#a78bfa"/>
      <path d="M13 31h18q8-23 16 0t16 0 16 0h16" fill="none" stroke="#c4b5fd" stroke-width="2"/>
      <circle cx="114" cy="24" r="14" fill="none" stroke="#a78bfa" stroke-width="3"/>
      <text x="155" y="19" fill="#c4b5fd" font-size="8.5" font-weight="700" text-anchor="middle">RF</text>
      <text x="155" y="32" fill="#8b98a5" font-size="7.2" text-anchor="middle">transceiver</text>
      <text x="89" y="48" fill="#8b98a5" font-size="7.2" text-anchor="middle">analog IP · passives · noise targets</text>
    </g>
    <text x="103" y="226" fill="#93c5fd" font-size="8" text-anchor="middle">own specification, differentiation, and market risk</text>
  </g>

  <!-- Design handoff is constrained by the foundry platform -->
  <path d="M230 177H275" stroke="#60a5fa" stroke-width="2.7" marker-end="url(#fmBlueArrow)"/>
  <g transform="translate(237 117)">
    <path d="M0 0h30l10 10v40H0Z" fill="#142b44" stroke="#60a5fa"/>
    <path d="M30 0v10h10" fill="none" stroke="#60a5fa"/>
    <g fill="#93c5fd" font-size="7.5" text-anchor="middle">
      <text x="20" y="23">GDS</text><text x="20" y="34">OASIS</text>
    </g>
  </g>
  <text x="253" y="190" fill="#93c5fd" font-size="8.2" text-anchor="middle">tapeout</text>
  <text x="253" y="202" fill="#8b98a5" font-size="7.5" text-anchor="middle">mask data + orders</text>

  <!-- Foundry: common process platform and capital assets -->
  <g transform="translate(281 78)">
    <text x="170" y="0" fill="#fbbf24" font-size="10.5" font-weight="700" text-anchor="middle">PURE-PLAY FOUNDRY</text>
    <path d="M0 15h340v253H0Z" fill="#17150f" stroke="#f59e0b" stroke-width="1.8" filter="url(#fmFabGlow)"/>

    <!-- Fab building / cleanroom -->
    <path d="M14 48h312v139H14Z" fill="#151b23" stroke="#52647b"/>
    <path d="M14 48l53-29h206l53 29" fill="#263548" stroke="#64748b"/>
    <g fill="#10233b" stroke="#60a5fa">
      <rect x="28" y="65" width="55" height="72" rx="5"/><rect x="93" y="65" width="55" height="72" rx="5"/>
      <rect x="158" y="65" width="55" height="72" rx="5"/><rect x="223" y="65" width="55" height="72" rx="5"/>
    </g>
    <g fill="#93c5fd" font-size="7.5" text-anchor="middle">
      <text x="55" y="82">DEPOSITION</text><text x="120" y="82">LITHOGRAPHY</text>
      <text x="185" y="82">ETCH / IMPLANT</text><text x="250" y="82">METROLOGY</text>
    </g>
    <!-- Physical equipment -->
    <g stroke="#60a5fa" fill="none">
      <path d="M41 119h29V91H41Z"/><circle cx="55" cy="103" r="7"/>
      <path d="M105 119h31V91h-31Z"/><path d="M111 98h19M111 105h19M111 112h19"/>
      <path d="M170 119h30V91h-30Z"/><path d="M176 97l18 16M194 97l-18 16"/>
      <path d="M235 119h30V91h-30Z"/><circle cx="250" cy="104" r="9"/>
    </g>

    <!-- Wafer lots move through hundreds of repeated steps -->
    <path d="M35 154H294" stroke="#f59e0b" stroke-width="2.5" marker-end="url(#fmGoldArrow)"/>
    <g fill="url(#fmWafer)" stroke="#c4b5fd">
      <ellipse cx="61" cy="154" rx="12" ry="5"/><ellipse cx="127" cy="154" rx="12" ry="5"/><ellipse cx="193" cy="154" rx="12" ry="5"/><ellipse cx="259" cy="154" rx="12" ry="5"/>
    </g>
    <text x="170" y="178" fill="#8b98a5" font-size="7.8" text-anchor="middle">process recipe + equipment matching + statistical control + yield learning</text>

    <!-- What the foundry pools -->
    <g transform="translate(14 197)">
      <rect width="312" height="53" rx="8" fill="#211b0d" stroke="#f59e0b"/>
      <g font-size="8" text-anchor="middle">
        <text x="52" y="15" fill="#fbbf24" font-weight="700">PROCESS IP</text>
        <text x="156" y="15" fill="#fbbf24" font-weight="700">CAPITAL</text>
        <text x="260" y="15" fill="#fbbf24" font-weight="700">CAPACITY</text>
        <text x="52" y="29" fill="#8b98a5">device models · rules</text>
        <text x="156" y="29" fill="#8b98a5">fab + tools + facilities</text>
        <text x="260" y="29" fill="#8b98a5">many wafer lots</text>
        <text x="156" y="44" fill="#e6edf3" font-size="8.3">shared platform amortizes fixed cost across customers</text>
      </g>
    </g>
  </g>

  <!-- Wafers and die return to the product companies -->
  <path d="M621 179H684" stroke="#34d399" stroke-width="2.7" marker-end="url(#fmGreenArrow)"/>
  <g transform="translate(650 97)">
    <text x="43" y="-8" fill="#6ee7b7" font-size="10.5" font-weight="700" text-anchor="middle">OUTPUT</text>
    <circle cx="43" cy="45" r="39" fill="url(#fmWafer)" stroke="#c4b5fd" stroke-width="2"/>
    <g stroke="#e9d5ff" stroke-width=".7" opacity=".8">
      <path d="M14 25h58M7 37h72M5 49h76M10 61h66M21 11v68M33 7v76M45 6v78M57 9v72M69 20v50"/>
    </g>
    <path d="M16 37h12v12H16Z" fill="#34d399" fill-opacity=".75" stroke="#6ee7b7"/>
    <path d="M45 49h12v12H45Z" fill="#34d399" fill-opacity=".75" stroke="#6ee7b7"/>
    <text x="43" y="96" fill="#8b98a5" font-size="7.8" text-anchor="middle">processed wafers</text>
    <g transform="translate(1 111)">
      <rect width="37" height="31" fill="#153127" stroke="#34d399"/>
      <rect x="47" width="37" height="31" fill="#153127" stroke="#34d399"/>
      <g fill="#6ee7b7" font-size="7" text-anchor="middle"><text x="18.5" y="18">KGD</text><text x="65.5" y="18">DIE</text></g>
      <text x="42" y="49" fill="#8b98a5" font-size="7.5" text-anchor="middle">test / known-good die</text>
    </g>
  </g>

  <!-- PDK and yield feedback close the engineering loop -->
  <path d="M432 346V376Q432 392 416 392H135Q119 392 119 376V335" fill="none" stroke="#a78bfa" stroke-width="2.5" stroke-dasharray="7 4" marker-end="url(#fmBlueArrow)"/>
  <g transform="translate(183 369)">
    <rect width="189" height="31" rx="15" fill="#211936" stroke="#a78bfa"/>
    <text x="94.5" y="13" fill="#c4b5fd" font-size="8.5" font-weight="700" text-anchor="middle">PDK + DESIGN RULES + MODELS</text>
    <text x="94.5" y="24" fill="#8b98a5" font-size="7.2" text-anchor="middle">manufacturing capability constrains design choices</text>
  </g>
  <path d="M692 259V408Q692 422 678 422H456Q442 422 442 408V347" fill="none" stroke="#34d399" stroke-width="2.3" marker-end="url(#fmGreenArrow)"/>
  <text x="574" y="416" fill="#6ee7b7" font-size="8.2" text-anchor="middle">wafer-sort data → yield learning → process improvement</text>

  <!-- Economic flywheel without duplicating the protected foundry chart -->
  <g transform="translate(24 420)">
    <path d="M0 0h421v22H0Z" fill="#10251f" stroke="#34d399"/>
    <g fill="#6ee7b7" font-size="8" font-weight="700" text-anchor="middle">
      <text x="55" y="14">CUSTOMER VOLUME</text><text x="161" y="14">UTILIZATION</text><text x="267" y="14">REINVESTMENT</text><text x="385" y="14">BETTER PDK / YIELD</text>
    </g>
    <g stroke="#34d399" stroke-width="1.8" marker-end="url(#fmGreenArrow)">
      <path d="M108 11h22"/><path d="M210 11h22"/><path d="M318 11h10"/>
    </g>
  </g>

  <text x="380" y="460" fill="#6b7684" font-size="11" text-anchor="middle">The model succeeds when specialization, confidentiality, predictable capacity, manufacturable design, and sustained yield reinforce one another.</text>
</svg>

Evaluation, roadmap discipline, and CFS connection. Foundry selection is a multi-year system decision. Teams compare actual PPA on representative blocks, SRAM and analog results, yield ramps, reticle and package strategy, mask and wafer cost, IP maturity, tool certifications, qualification, and recovery plans. “Never compete with customers” is an important pure-play principle, but execution quality ultimately depends on predictable manufacturing and transparent technical collaboration. Due diligence separates measured facts from marketing categories and forward-looking plans. Check the date, product form factor, memory configuration, power limit, software release, process variant, package, and whether a number is peak, typical, estimated, or independently reproduced. Company revenue rankings and foundry shares move with cycles, currency, reporting boundaries, and whether wafer manufacturing or end-product sales are counted. Procurement adds total landed cost, supply assurance, licensing terms, support, lifecycle, compliance, and exit options. Engineering teams should preserve traceable assumptions and revisit them when a roadmap, regulation, yield curve, or workload changes. CFS connects this topic to semiconductor architecture, implementation, verification, manufacturing, packaging, test, and deployed AI-system tradeoffs across the platform.