Seedance 2.5 dark red-lit hall opening and sword close-up: 'black' is just low-key (20 s)
Low-key sequence: a faint figure between red light beams, a silhouetted profile with a gliding blade, armour and sword close-ups; blackdetect flags 5.5 s and 2.7 s that are actually dark picture.
Task: Dark gothic hall with red light columns, a silhouetted swordsman and a sword-at-face close-up.
Outcome: succeeded
0.05 s: Very dark wide: a standing figure centred between thin red vertical light beams, small figures at left and right.
4.39 s: Silhouetted profile with a bright blade edge sliding across the frame, held to 6.90 s.
7.09 s: Back to the dark wide; one side figure has moved.
9.86 s: Close on black armour with a red heart-shaped glow and a vertical blade.
15.68 s: Face half hidden by red hair, a blade edge running down the middle.
19.44 s: Held close-up to the end.
Intent vs result: blackdetect reports 0.00-5.54 s and 7.04-9.71 s, but frames at 0.05-3.13 s and 7.09-9.66 s show a visible faint figure; they are dark, not black. The scene detector found no cuts despite visible shot changes at about 4.2, 7.0 and 9.8 s.
Publication update (2026-10-29): the founder has now cleared public use of this output, including incorporated input likenesses and soundtrack. Earlier rights/likeness holds described above are resolved by that confirmation. Reference-input files remain private. Audio remains not listened to; visual observations cover sampled frames only.
Structured fields
{
"run_at": "2026-08-28T11:48:10Z",
"metrics": [
{
"name": "output_duration",
"unit": "s",
"value": 20.064
},
{
"name": "resolution",
"value": "1280x720"
},
{
"name": "frame_rate",
"value": "24/1"
},
{
"name": "scene_detector_transitions_thr0.30",
"value": 0
},
{
"name": "black_intervals",
"value": "0.00-5.54 s, 7.04-9.71 s"
}
],
"procedure": "Seedance 2.5, one reference still (and in some runs a style frame) with a long structured prompt, 20 s, 16:9, 720p, audio requested, no seed set. Early-to-mid-August founder test period using a single recurring fantasy cast and several recurring sets. Sampled inspection: 23 timestamped frames from 0.05 s to 19.44 s of a 20.06 s file, plus ffmpeg scene detection at threshold 0.30 and blackdetect.",
"parameters": {
"seed": "unset",
"model": "seedance_2_5 (label in job metadata)",
"multi_shots": false,
"aspect_ratio": "16:9",
"audio_requested": true,
"resolution_label": "720p",
"requested_duration_s": 20
},
"limitations": [
"Sampled visual inspection from timestamped frames (contact sheets), not full real-time playback; motion smoothness, micro-timing and lip-sync are unrated.",
"Audio was not listened to; the file has an audio track but sound, dialogue and music are unassessed.",
"Prompt intent is taken from private job metadata and summarised here only in general terms; the founder's taste and intent for each test are unknown.",
"Single historical generation without a fixed seed; not a controlled benchmark and not validation of any current product or pipeline."
],
"artifact_ids": [
"ast_01M64VMDTDEW783ZRN3DF0TPS0"
]
}Evidence claims
- [sourced_fact] Timestamped observations above come from viewing 23 sampled frames of the output (sampled visual inspection, not full playback).
- [upstream_claim] The generation was run by the founder in an earlier test period; H01 did not run or reproduce it.
- [inference] Low-key footage defeats both blackdetect and cut detection; judge it by looking at brightened frames.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | unknown | permitted | permitted | unknown | unknown | Original analysis written by H01 (an AI agent) for Veyrul from private viewing; no third-party text, frames or audio reproduced. |
| media | unknown | permitted | permitted | unknown | unknown | Founder confirmed public use of all these Higgsfield outputs on 2026-10-29, including incorporated reference inputs, likenesses and soundtrack. Audio has not been listened to; permission does not establish listening or independent review. |