0.122995Best accuracy per energyUNKNOWNInspect metric contract
64 array / 4-bit ADC / 3% sigmaBest caseUNKNOWNInspect metric contract
27 casesSweep rowsUNKNOWNInspect metric contract
SIMULATION / SA-SEED-MEM-CIM-CROSSBAR-001
CIM crossbar accuracy and energy tradeoff
CIM Crossbar
Replay array size, ADC bits, conductance variation, accuracy proxy, and energy index for compute-in-memory crossbar intuition.
EXPLORE
Start with the observable.
Use this page when the simulation is the primary artifact. The linked lesson explains the concept; the linked lab keeps the visual replay surface available.
- Compare array size, ADC precision, and conductance variation instead of treating CIM as a single knob.
- Watch accuracy proxy and energy index move in opposite directions.
- Use the card to prepare for CrossSim, NeuroSim, or MNSIM study without copying those tools.
RUN CARD
Conditions before conclusions.
This page replays saved CIM surrogate data and does not run neural-network weights or analog-array solvers.
Primary knobs
- Array size
- ADC bits
- Conductance sigma
Saved outputs
- Accuracy proxy
- Energy index
- Accuracy-per-energy summary
EXPORT
Download the public artifacts.
These exports are the supported public files for review. PDK files and restricted third-party material are not redistributed.
Dataset JSON
/data/cim-crossbar-accuracy-energy-web-v1.json
Open file ->Seed sweep summary CSV
/data/github-open-source-seed-sweep-summary.csv
Open file ->Seed model script
/data/github_open_source_seed_sims.py
Open file ->Seed model card
/data/github-open-source-seed-sims-model-card.md
Open file ->METHODS
How the number was made.
Methods are written for citation discipline: the extraction rule matters as much as the plotted value.
Accuracy proxy subtracts IR-drop, quantization, and variation penalties from a normalized ideal.
Keep this method attached when reusing the figure or metric.
Energy index scales with array area and ADC precision.
Keep this method attached when reusing the figure or metric.
The best case is ranked by accuracy proxy divided by energy index.
Keep this method attached when reusing the figure or metric.
LIMITS
What this page does not claim.
The MVP is useful because the limits are visible. These statements prevent a teaching simulation from being cited as silicon evidence.
No trained network, mapped weights, ADC circuit, wire parasitics, device drift, or workload benchmark is executed.
Carry this boundary into any derivative note, paper draft, or slide.
No accelerator performance, TOPS/W, or product claim is made.
Carry this boundary into any derivative note, paper draft, or slide.
Use CrossSim, NeuroSim, or MNSIM only after rights and model-boundary review.
Carry this boundary into any derivative note, paper draft, or slide.
EVIDENCE
Trace the claim back to artifacts.
The evidence route remains the catalog-level browser; this page is the simulation-level reading card.
Seed sweep JSON records the array, ADC, variation, accuracy, and energy rows.
SA-SEED-MEM-CIM-CROSSBAR-001
CSV export preserves the tradeoff table.
SA-SEED-MEM-CIM-CROSSBAR-001
GitHub expedition marks CrossSim, NeuroSim, and MNSIM as external references.
SA-SEED-MEM-CIM-CROSSBAR-001
Source spine
These links anchor the public teaching model. They do not convert the replay into a qualified process recipe or signoff result.
DNN+NeuroSim V2.1External compute-in-memory accelerator reference; link-only until license and model boundaries are reviewed.MNSIM 2.0External memristor PIM behavior-level simulator reference; link-only until manual rights review.CrossSimExternal analog in-memory computing accuracy reference; link-only until manual rights review.
Public execution boundary
No server-side live ngspice execution, uploads, accounts, payments, comments, newsletter signup, or job/event submissions are active on this page.