← Back to Chip Foundry Services

Glossary

15,747 technical terms and definitions

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Showing page 43 of 315 (15,747 entries)

cluster tool module availability prediction machine learning

cluster tool module availability prediction, cluster tool module, tool module availability, module availability prediction, cluster tool, tool module, module availability

Cluster Tool Module Availability Prediction with Machine Learning Introduction & Motivation Cluster Tool Module Availability Prediction with Machine Learning addresses a central…

clustering-based-anomaly machine learning

Clustering-Based Anomaly Detection Introduction & Motivation Clustering-Based Anomaly Detection addresses a central problem in anomaly detection and (target) scoring: how…

clustering-based-transfer machine learning

Clustering-Based Transfer Learning Introduction & Motivation Clustering-Based Transfer Learning addresses a central problem in transfer learning and (target) adaptation: how…

clustering-deep-features machine learning

Deep Feature Clustering Introduction & Motivation Deep Feature Clustering addresses a central problem in vision processing and cluster-quality-metric prediction: how to Build…

clustering-time-series-unsupervised machine learning

clustering time series unsupervised, clustering time series, time series unsupervised, clustering time, time series, series unsupervised

Clustering Time Series Unsupervised Introduction & Motivation Clustering Time Series Unsupervised addresses a central problem in timeseries process control…

cml current mode logic swing prediction machine learning

cml (current mode logic) swing prediction, cml (current mode, (current mode logic), mode logic) swing, logic) swing prediction, cml (current, (current mode, mode logic)

CML (Current Mode Logic) Swing Prediction with Machine Learning Introduction & Motivation CML (Current Mode Logic) Swing Prediction with Machine Learning addresses a central…

cmos 1963 two opposing transistors solve power problem

fairchild 1963 wanlass sah complementary pair logic, cmos transistor 1963 exactly one conducts one blocks, silicon cmos 1963 no steady path supply to ground

Why Two Opposing Transistors Solve the Power Problem 1962 Closed On: Zero Current, Not Just Zero Gate Current 1. Why a Near-Zero Gate Current Was Never the Whole Answer This step…

cmos 1963 build well host one type inside other

fairchild 1963 well doping opposite substrate deep diffusion, cmos transistor 1963 two device types one piece silicon, silicon cmos 1963 well depth must clear source drain

Build a Well: Hosting One Transistor Type Inside a Substrate Doped for the Other, Reversed Locally Rather Than Across the Whole Wafer 1. Why One Wafer Now Has to Carry Two…

cmos 1963 diffuse well to the depth devices need

fairchild 1963 lateral spread well much larger scale now, cmos transistor 1963 deep well diffusion costs chip area, silicon cmos 1963 well proximity minimum spacing rule

Diffuse the Well to the Depth Devices Actually Need: The Same Lateral-Spread Fact, a Much Bigger Number This Time 1. Why a Deep Diffusion Costs Far More Chip Area Than a Shallow…

cmos 1963 source drain both transistor types two diffusions

fairchild 1963 nmos pmos diffusion must never cross well edge, cmos transistor 1963 two separate masking sequences one wafer, silicon cmos 1963 wrong dopant crossing well boundary risk

Define Source and Drain for Both Transistor Types: Two Diffusions That Must Never Touch Each Other's Territory 1. Why This Wafer Needs Two Masking Sequences Where Every Earlier…

cmos 1963 wire pair first two transistor logic gate

fairchild 1963 complementary inverter no resistor needed, cmos transistor 1963 eliminates the resistor 1960 needed, silicon cmos 1963 metal ties drains together gates together

Wire the Pair Into This Project's First Two-Transistor Logic Gate: The Resistor 1960 Had to Size, This Gate Doesn't Need at All 1. Why This Gate Needs No Component This Project…

cmos 1963 verify near zero static current both states

fairchild 1963 subthreshold leakage sets the real floor, cmos transistor 1963 first gate level current measurement, silicon cmos 1963 measured current orders of magnitude lower

Verify Near-Zero Static Current in Both Logic States: Not Exactly Zero, But Orders of Magnitude Away From the Alternative 1. Why Step 1's Equation Finally Needs a Number Behind It…

cmos 1963 parasitic thyristor hiding inside every pair

fairchild 1963 cross coupled pnp npn regenerative feedback, cmos transistor 1963 well creates its own failure mode, silicon cmos 1963 latch low impedance shorts supply ground

Confront a Failure Mode the Well Itself Makes Possible: A Parasitic Thyristor Hiding Inside Every Complementary Pair 1. Why the Same Structure That Enables This Series Also…

cmos 1963 measure switching power delay tradeoff

fairchild 1963 dynamic power scales with switching frequency, cmos transistor 1963 static advantage not unconditional, silicon cmos 1963 crossover frequency dynamic dominates static

Measure Switching Power and the Power-Delay Trade-Off: The Power Step 6 Never Claimed to Eliminate 1. Why Near-Zero Static Power Was Never the Whole Picture This step measures…

cmos 1963 qualify against single channel mosfet

fairchild 1963 double transistor count extra well area, cmos transistor 1963 more steps for power not speed, silicon cmos 1963 area cost static power advantage trade

Qualify Against the Single-Channel MOSFET Just Completed: More Area, More Steps, a Power Advantage 1962 Could Never Offer 1. Why This Series' Own Gate Costs More to Build Before…

cmos 1963 classify release architecture digital logic standard

fairchild 1963 fixed area cost recurring power cost compound, cmos transistor 1963 gate count under power budget finally real, silicon cmos 1963 completes figure of merit 1962 could not

Classify and Release: The Architecture Digital Logic Will Standardize On 1. Why This Series' Figure of Merit Finally Delivers What 1962's Own Could Not This closing step…

cmos image sensor dark current drift prediction machine learning

cmos image sensor dark current drift prediction, cmos image sensor, image sensor dark, sensor dark current, dark current drift, current drift prediction, cmos image

CMOS Image Sensor Dark Current Drift Prediction with Machine Learning Introduction & Motivation CMOS Image Sensor Dark Current Drift Prediction with Machine Learning addresses…

cmos image sensor microlens alignment prediction with machine learning

cmos image sensor microlens alignment prediction machine learning, cmos image, sensor microlens alignment, cmos, cmos image sensor microlens alignment prediction

CMOS Image Sensor Microlens Alignment Prediction with Machine Learning Introduction & Motivation CMOS Image Sensor Microlens Alignment Prediction with Machine Learning addresses…

cmos image sensor pixel defect prediction with machine learning

cmos image sensor pixel defect prediction machine learning, cmos image, sensor pixel defect, cmos, cmos image sensor pixel defect prediction, cmos image sensor, image sensor

CMOS Image Sensor Pixel Defect Prediction with Machine Learning Introduction & Motivation CMOS Image Sensor Pixel Defect Prediction with Machine Learning addresses a central…

cmos integration schemes

process integration, gate-first gate-last, strain engineering, FinFET nanosheet GAA CFET, device architecture evolution, replacement metal gate, standard cell scaling, foundry process complexity, node transition

CMOS integration schemes are the overall architectural strategies that define how complementary NMOS and PMOS transistors actually get built together on the same substrate…

cmos logic 1980s confront the parasitic thyristor failure mode

cmos logic 1980s step 7 confront the parasitic thyristor failure mode, confront the parasitic thyristor failure mode

Confront a Failure Mode the Well Itself Makes Possible: A Parasitic Thyristor Hiding Inside Every Complementary Pair 1. Why the Same Structure That Enables This Series Also…

cmos logic 1980s define source and drain for both transistor types

cmos logic 1980s step 4 define source and drain for both transistor types, define source and drain for both transistor types

Define Source and Drain for Both Transistor Types: Two Diffusions That Must Never Touch Each Other's Territory 1. Why This Wafer Needs Two Masking Sequences Where Every Earlier…

cmos logic 1980s qualify against single channel mosfet

cmos logic 1980s step 9 qualify against single channel mosfet, qualify against single channel mosfet

Qualify Against the Single-Channel MOSFET Just Completed: More Area, More Steps, a Power Advantage 1962 Could Never Offer 1. Why This Series' Own Gate Costs More to Build Before…

cmos logic 1980s size decoupling capacitance against the switching noise budget

cmos logic 1980s step 15 size decoupling capacitance against the switching noise budget, size decoupling capacitance against the switching noise budget

Size a Local Reservoir of Charge So the Grid Never Has to Answer Instantly 1. Separate What the Grid Can Supply Quickly From What a Transition Demands Instantly Step 14 showed…

cmos logic 1980s test the margin against hmos ii specifically

cmos logic 1980s step 17 test the margin against hmos ii specifically, test the margin against hmos ii specifically

Run the Harder Test: Does the Margin Survive Against NMOS's Best Possible Answer 1. Substitute HMOS-II's Numbers for First-Generation HMOS's Step 16 confirmed CMOS's power…

cmos logic 1980s wire the pair into the first two transistor gate

cmos logic 1980s step 5 wire the pair into the first two transistor gate, wire the pair into the first two transistor gate

Wire the Pair Into This Project's First Two-Transistor Logic Gate: The Resistor 1960 Had to Size, This Gate Doesn't Need at All 1. Why This Gate Needs No Component This Project…

cmos crossover 1980 twin well latchup density penalty

twin well cmos independent n well p well tuning, parasitic pnp npn thyristor latchup loop gain beta, guard ring substrate tie layout area cost latchup prevention

Build Two Wells on One Substrate, and Confront the Parasitic Transistor Density Reintroduces 1. State What Housing Both Device Types on One Substrate Actually Requires…

cmos crossover 1980 test margin against hmos ii specifically

hmos ii static term unchanged by scaling i_dd structural, cmos margin narrows numerically widens structurally scaling headroom, nmos physical limit versus cmos still has room to scale

Run the Harder Test: Does the Margin Survive Against NMOS's Best Possible Answer 1. Substitute HMOS-II's Numbers for First-Generation HMOS's Step 6 confirmed CMOS's power…

cmp carrier film membrane fatigue prediction machine learning

cmp carrier film membrane fatigue prediction, cmp carrier film, carrier film membrane, film membrane fatigue, membrane fatigue prediction, cmp carrier, carrier film

CMP Carrier Film Membrane Fatigue Prediction with Machine Learning Introduction & Motivation CMP Carrier Film Membrane Fatigue Prediction with Machine Learning addresses a central…

cmp carrier retaining ring degradation prediction machine learning

cmp carrier retaining ring degradation prediction, cmp carrier retaining, carrier retaining ring, retaining ring degradation, ring degradation prediction

CMP Carrier Retaining Ring Degradation Prediction with Machine Learning Introduction & Motivation CMP Carrier Retaining Ring Degradation Prediction with Machine Learning addresses…

cmp conditioner disk wear prediction with machine learning

cmp conditioner disk wear prediction machine learning, cmp conditioner, disk wear prediction, cmp conditioner disk wear prediction, cmp conditioner disk, conditioner disk wear

CMP Conditioner Disk Wear Prediction with Machine Learning Introduction & Motivation CMP Conditioner Disk Wear Prediction with Machine Learning addresses a central problem in CMP…

cmp diamond disk wear life prediction machine learning

cmp diamond disk wear life prediction, cmp diamond disk, diamond disk wear, disk wear life, wear life prediction, cmp diamond, diamond disk, disk wear, wear life, life prediction

CMP Diamond Disk Wear Life Prediction with Machine Learning Introduction & Motivation CMP Diamond Disk Wear Life Prediction with Machine Learning addresses a central problem…

cmp dishing erosion prediction machine learning

cmp dishing and erosion prediction, cmp dishing erosion, dishing erosion prediction, cmp dishing, dishing erosion, erosion prediction

CMP Dishing and Erosion Prediction with Machine Learning Introduction & Motivation CMP Dishing and Erosion Prediction with Machine Learning addresses a central problem in copper…

cmp downforce zone control drift prediction with machine learning

cmp downforce zone control drift prediction machine learning, cmp downforce, zone control drift, cmp downforce zone control drift prediction, cmp downforce zone

CMP Downforce Zone Control Drift Prediction with Machine Learning Introduction & Motivation CMP Downforce Zone Control Drift Prediction with Machine Learning addresses a central…

cmp endpoint detection and overpolish prevention with machine learning

cmp endpoint detection overpolish prevention machine learning, cmp endpoint, detection overpolish prevention, cmp rr, cmp endpoint detection and overpolish prevention

CMP Endpoint Detection and Overpolish Prevention with Machine Learning Introduction & Motivation CMP Endpoint Detection and Overpolish Prevention with Machine Learning addresses…

cmp endpoint detection signal drift prediction machine learning

cmp endpoint detection signal drift prediction, cmp endpoint detection, endpoint detection signal, detection signal drift, signal drift prediction, cmp endpoint

CMP Endpoint Detection Signal Drift Prediction with Machine Learning Introduction & Motivation CMP Endpoint Detection Signal Drift Prediction with Machine Learning addresses…

CMP modeling

chemical mechanical polishing, CMP simulation, planarization, dishing, erosion

Chemical Mechanical Planarization (CMP) Modeling in Semiconductor Manufacturing 1. Fundamentals of CMP 1.1 Definition and Principle Chemical Mechanical Planarization (CMP) is…

cmp pad asperity wear rate prediction machine learning

cmp pad asperity wear rate prediction, cmp pad asperity, pad asperity wear, asperity wear rate, wear rate prediction, cmp pad, pad asperity, asperity wear, wear rate

CMP Pad Asperity Wear Rate Prediction with Machine Learning Introduction & Motivation CMP Pad Asperity Wear Rate Prediction with Machine Learning addresses a central problem…

cmp pad conditioning and slurry flow optimization with machine learning

cmp pad conditioning slurry flow optimization machine learning, cmp pad, conditioning slurry flow, cmp nu, cmp pad conditioning and slurry flow optimization

CMP Pad Conditioning and Slurry Flow Optimization with Machine Learning Introduction & Motivation CMP Pad Conditioning and Slurry Flow Optimization with Machine Learning addresses…

cmp pad conditioning effectiveness prediction machine learning

cmp pad conditioning effectiveness prediction, cmp pad conditioning, pad conditioning effectiveness, conditioning effectiveness prediction, cmp pad, pad conditioning

CMP Pad Conditioning Effectiveness Prediction with Machine Learning Introduction & Motivation CMP Pad Conditioning Effectiveness Prediction with Machine Learning addresses…