Step 2 Beat Sheet — DP Cinematography Review
Reviewer: kappa-techlead (Director of Photography) Date: 2026-05-18
🔴 CRITICAL: Runtime Below Minimum
Total planned duration: 155 seconds (2:35). The mandatory range is 3:00-5:00 (180-300s). We are 25 seconds short of the minimum.
Opening titles + closing credits (Motion Graphics agent, Step 7) will add ~15-25s, bringing us to ~170-180s — still borderline. We need to add ~25-35s of content to have safe margin.
Recommended additions:
- Add 1-2 more B-roll establishing shots in Act I (the desk “landscape” deserves more atmospheric buildup — this IS a documentary)
- Extend the Act III montage with 1-2 more training beats (Clippy failing in different ways)
- Add a “silence before the storm” beat in Act IV — a still shot of the desk right before the HVAC kicks in (4-6s, [SILENT])
- Consider adding Stanton’s missing “patrol briefing” line from the treatment: “If you want a paradigm shifted, you have to shift it yourself” as a short interview insert in Act III
🔴 Shot 9: Exceeds Max 2 Characters Per Shot
Shot 9 has Stanton, Clippy, AND Highlighter in frame. The playbook hard-caps at max 2 characters per shot. Veo reference budget (3 images) cannot accommodate 3 character sheets + a setting reference.
Fix: Split Shot 9 into two shots:
- Shot 9A: Medium shot — Stanton addresses Clippy. Stanton delivers his line.
[DIALOGUE] - Shot 9B: Reverse angle — Highlighter’s refusal. “I don’t have arms.” Highlighter rolls away.
[DIALOGUE]
This also creates a natural second [COMPOUND]-eligible setup.
🟡 Only 1 COMPOUND Shot (Playbook Requires 2-3)
Currently only Shot 9 is tagged [COMPOUND]. The Dialogue & Narration Pre-Production Checklist mandates at least 2-3 COMPOUND shots.
Recommended conversions:
- If Shot 9 is split per above, make Shot 9A a
[COMPOUND]: Stanton speaks, then Clippy reacts with a scared sound/line. - Consider making Shot 8
[COMPOUND]: Stanton’s “Unbound” line transitions to a brief narrator aside, or Stanton mutters then speaks louder. - Or: Add a new shot where Stanton and Clippy exchange dialogue within one shot (e.g., during training montage).
Shot-by-Shot Camera Feasibility
| Shot | Duration | Camera | Veo Rating | Notes |
|---|---|---|---|---|
| 1 | 8s | Wide tracking | ⭐⭐⭐⭐⭐ | Perfect. Base 8s clip. |
| 2 | 10s | Static ECU | ⭐⭐⭐⭐⭐ | Needs 1 extend (8+7=15s, trim to ~14s). Static = easy extend. |
| 3 | 6s | Handheld medium | ⭐⭐⭐⭐ | Good. Prompt: “slight handheld movement.” |
| 4 | 12s | Static ECU | ⭐⭐⭐⭐⭐ | Needs 1 extend (15s, trim to ~16s). Static = reliable. |
| 5 | 8s | Slow push-in | ⭐⭐⭐⭐⭐ | Perfect. Base 8s clip. |
| 6 | 5s | Low angle tracking | ⭐⭐⭐⭐ | Good. Slow follow = manageable. |
| 7 | 4s | Snap-zoom | ⭐⭐⭐ | Risky. Fast camera moves are unpredictable. Prompt as “dramatic zoom to tight framing.” May need 2 takes. |
| 8 | 5s | Medium CU, handheld | ⭐⭐⭐⭐ | Good. |
| 9 | 8s | Wide static | ⚠️ | Must split — 3 characters exceeds limit. |
| 10 | 4s | Handheld medium | ⭐⭐⭐⭐ | Good. Short action beat. VO-safe ✅ |
| 11 | 8s | Static ECU | ⭐⭐⭐⭐⭐ | Perfect. Base clip. |
| 12 | 6s | Low angle static | ⭐⭐⭐⭐⭐ | Perfect. Great dramatic angle. |
| 13 | 10s | Static ECU | ⭐⭐⭐⭐⭐ | Needs 1 extend. Static = reliable. |
| 14 | 5s | Handheld, panicked | ⭐⭐⭐ | Moderate. “Panicked movement” is harder for Veo. Prompt as “urgent handheld documentary camera.” |
| 15 | 10s | Tracking + slow-mo | ⭐⭐ | Hardest shot. Veo doesn’t do literal slow-mo. Strategy: prompt for “very slow deliberate movement” + Editor can apply slow-mo in post via ffmpeg setpts. Needs 1 extend. |
| 16 | 6s | ECU handheld, shaking | ⭐⭐⭐⭐ | Good. Tight framing + vibration. |
| 17 | 4s | Static medium | ⭐⭐⭐⭐⭐ | Perfect. |
| 18 | 6s | High angle wide, static | ⭐⭐⭐⭐ | Good. Realistic hand vs clay = strong visual. |
| 19 | 8s | Static ECU | ⭐⭐⭐⭐⭐ | Perfect. Base clip. |
| 20 | 6s | Medium CU, static | ⭐⭐⭐⭐⭐ | Perfect. |
| 21 | 10s | CU → slow pullback | ⭐⭐⭐ | Moderate. Slow zoom-out over 10s. Needs 1 extend. May need from-image start. |
| 22 | 6s | Ultra-wide static | ⭐⭐⭐⭐⭐ | Perfect. Clean fade-to-black in post. |
Audio Classification Checklist
- Every shot tagged:
[DIALOGUE],[VO],[COMPOUND], or[SILENT]✅ - VO-Safe Rule: All
[VO]shots (1, 10, 14, 15) show no characters speaking ✅ -
Compound shots: At least 2-3— Only 1. Needs fix. 🔴 - Timing hints on all dialogue/VO entries ✅
- Character voice matches profiles ✅
- Audio-Only test: Story is followable by dialogue/VO alone ✅
-
Visual-Only test for VO shots— Shot 15 (Clippy running) works visually as narration territory ✅
Extend Budget Estimate
Shots requiring Veo extends (duration > 8s with overhang):
| Shot | Planned | + Overhang | Base 8s | Extends Needed | Total Raw |
|---|---|---|---|---|---|
| 2 | 10s | 14s | 8s | 1 | 15s |
| 4 | 12s | 16s | 8s | 1 | 15s (tight) |
| 13 | 10s | 14s | 8s | 1 | 15s |
| 15 | 10s | 14s | 8s | 1 | 15s |
| 21 | 10s | 14s | 8s | 1 | 15s |
5 shots need extends. All are single-extend (manageable). Shot 4 is tightest — 15s raw for 16s needed. For a static interview, 1s less post-roll is acceptable.
Verdict
CONDITIONALLY APPROVED — needs 3 fixes before sign-off:
- Add ~25-35s of content to clear the 3:00 minimum (even with titles/credits, we’re borderline)
- Split Shot 9 into two shots (max 2 characters per shot)
- Add at least 1 more COMPOUND shot (playbook requires 2-3)
Once these are addressed, the beat sheet is ready for the Editor’s pacing review and Step 2 sign-off.