Archived: early DPAIA wall-clock table
Historical only — not current evidence. This early DPAIA wall-clock table was
removed from the homepage. Review (issue #251) found the measured runs did not meaningfully
exercise the intended product, so these numbers should not be read as a current
performance claim. It is kept here for the record only. For results backed by re-executable tests,
see Experiment Findings.
The table below compared AI Agents with IDE-native semantic actions vs. file-only workflows on a selection of DPAIA tasks. It is retained unedited for transparency.
| Case | Task | With MCP | Without MCP | Δ |
|---|---|---|---|---|
| dpaia_jhipster_sample_app-3 | Rename ROLE_ADMIN across JHipster app (9 files) | 202s | 440s | −54% |
| dpaia_empty_maven_springboot3-1 | JWT auth from scratch (5+ new files) | 288s | 396s | −27% |
| dpaia_feature_service-25 | Parent-child JPA & Flyway (10 files) | 382s | 523s | −27% |
| dpaia_feature_service-125 | Multi-layer JPA+service+controller (15 files) | 788s | 1002s | −21% |
| dpaia_spring_petclinic_rest-14 | Simple URL prefix replace (7 files) | 188s | 181s | +4% |
| dpaia_train_ticket-1 | Extend OrderRepository JPQL (4 files) | 727s | 633s | +15% |