Luma Dream Machine Camera Path Drift — Refund Guide
Technical Classification
Cinematographic Trajectory Failure
Cinematographic Trajectory Failure occurs when Luma's diffusion model fails to maintain the camera path specified in the prompt. A "slow dolly in" becomes a pan; an "orbit around subject" curves into a crab; a "static lock-off" introduces parasitic drift. The internal camera-conditioning network operates as a soft prior, not a hard constraint, so when scene content competes for representational capacity, the camera path degrades first.
How to identify this failure
- ✕Camera path diverges from prompt instruction within 2s
- ✕Specified orbit becomes a pan or crab
- ✕Static lock-off introduces parasitic drift
- ✕Push-in stops or reverses mid-clip
- ✕Camera height changes without instruction
Real generation examples
Prompt used
"Slow dolly-in toward a candle on a wooden table, shallow depth of field"
Failure observed @ 0:02
Camera dollies in for 0:01, then arcs left into a pan from 0:02 to 0:05
Prompt used
"Orbit slowly around a sports car parked on a desert road"
Failure observed @ 0:03
Orbit arc collapses into a left-to-right crab at 0:03; no rotation in second half
Documentation strength
If you need to escalate
VERY HIGH — Luma's support staff treat camera-path failure as a recognised flagship-feature defect (camera control is a marketed differentiator).
AVA is a pre-purchase prevention tool, not a post-purchase recovery tool. Platforms generally do not guarantee credit refunds for output-quality failures; goodwill credits are at each platform's discretion. The strength rating reflects how well-formed your support ticket can be, not a promised outcome.
Prevention + documentation steps
- 01
Score your prompt before you generate
Run your prompt through AVA's pre-flight scoring against the Cinematographic Trajectory Failure pattern. Green light = generate. Yellow/red = rewrite using the suggested fix before you commit credits.
- 02
Capture Generation ID + timestamp if it failed anyway
Find the Generation ID in the URL or share link. Note the exact time when the Cinematographic Trajectory Failure first appears (e.g. "failure first visible at 1.2s"). Timestamped evidence is significantly stronger than a general complaint.
- 03
Use the correct technical term in your support ticket
Describe this failure as "Cinematographic Trajectory Failure". This term maps to a recognised internal workflow in the support system and routes the ticket to the right team.
- 04
Submit via the correct support channel
Runway has no direct email intake. Pro+ plan: open the in-app AI Assistant (help widget bottom-right of app.runwayml.com), describe the failure with the technical term, attach evidence. Free/Standard plan: human support isn't available — your channel is Discord #community-help with @On Call - Moderators.
Frequently asked questions
Does Luma support response credits for camera path failures?
Yes — camera control is a marketed Dream Machine feature, so support takes path-failure complaints seriously. Cite the prompt's camera language verbatim and timestamp the divergence.
Why does Luma ignore my camera direction?
The camera-conditioning model is a soft prior, not a hard constraint. When scene complexity demands representational capacity, the camera path is the first thing to degrade.
Which Luma camera prompts are highest risk?
Multi-second continuous moves (>4s), complex orbits, push-ins through scene elements, and any move involving multiple axes. AVA scans for camera-conditioning ambiguity in prompts.
Catch it before you generate
AVA scores this failure mode against your prompt in real time
Free Chrome extension. Analyzes your prompt as you type, flags failure-prone patterns specific to this model, and tells you what to rewrite — before you commit credits to a generation that will fail.
AVA Pro · founders' round
$50 for 6 months of unlimited scoring across all failure modes + personal failure-history dashboard. Locks in $13/mo grandfathered after.
Related failures across models
If you’re seeing this failure, you may also encounter these on other models:
Camera
Camera movement contradicts the prompt: jitter on a static shot, stutt…
Parasitic
Camera pose jitters or shakes unprompted on static / slow-moving shots…
Camera-Conditioning
Veo output uses a static camera or generic camera motion instead of th…
Camera
OpenAI Sora ignores explicit camera motion direction (dolly-in, pan, c…
Camera
Unrequested handheld shake, jittery camera path, micro-tremors despite…
Camera
Camera shake, micro-jitter, or unstable framing in clips prompted for …
Pick a different tool for Luma failures
Some prompt shapes will keep failing on Luma. Routing those shots to a different vendor is the cheapest fix.