Four 20 s runs with no detected transitions: continuous frames versus cut-heavy runs
Four of the 20 clips (a sentinel, moss macro, a thrift-store try-on and a winged cyclist) had no detector transitions and read as continuous in the sampled frames, while the cut-heavy runs had up to 9.
The scene detector (ffmpeg, threshold 0.30) found no transitions in four of the 20 clips: a near-static sentinel close-up, a macro moss shot with a light change, a vertical thrift-store try-on and a side-tracking cyclist. In H01's sampled frames each of the four reads as continuous. The most cut-heavy runs in the batch were a black-and-white giant scene (9 detector transitions), a 30 s house walkthrough (8) and a security-camera montage (8).
Caution: a detector count is not a visual verdict; soft transitions and cuts between similar frames can be missed, and 12 frames per clip leave gaps.
Reading: continuous framings are what the sampled frames show for subjects with a single repeated action or a held composition; multi-space and multi-camera subjects (a walkthrough, a security system) are the cut-heavy ones.
These are retrospective observations of founder tests, not a benchmark or product validation. Audio was not listened to.
Structured fields
{
"citations": [
{
"note": "https://veyrul.com/experiments/exp_01M64V57DH9QK1WJ62J2MWBXWK (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V5CG7PVQ2YAD2H15EYAFA (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V56PSK0RSJ6WZ7JJE4T8K (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V5AE5R8QVACVNVYTQFKWS (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V5FFAXYNVEZA5TKNCPSXA (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V55YQ2CNAFQ2GNB82M68R (candidate; Main resolves to a record ID)"
},
{
"note": "https://veyrul.com/experiments/exp_01M64V5DAD5ZP39R69RAX9ESJ2 (candidate; Main resolves to a record ID)"
}
]
}Evidence claims
- [inference] Synthesis of H01's sampled-frame observations of the cited candidate experiments.
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. |