PARALLAX METROLOGY VTL STUDIO
STUDIO RECORD / PUBLIC EDITION

Visual stages, comparisons and decisions from the original run. Internal prompts, operating documents and downloadable implementation files are omitted. Some interactive comparisons are shown as separate images.

VTL Studio / Reach–Retreat / another native sitting

Let the reach and weight lead.

This round works on the existing drawing. The question is how to quiet the clothing without making the figure inert. Three native alternatives were compared; the retained mass received nine selective arm and ground accents.

Retained as a working revision. The clothing is less insistent. The larger openings failed. This is a clearer grouping of values, not a claim that the drawing is finished or that it has found a new identity.
Selected: quieter clothed mass, then selective native accents. 580 new retained commands; 3,896 in the complete replayable history.
Selected: quieter clothed mass, then selective native accents. 580 new retained commands; 3,896 in the complete replayable history.

Full-resolution drawing · · ·

Before → after

Before: bright clothing grooves compete with the gesture.
Before: bright clothing grooves compete with the gesture.
After: the hip and thigh read more continuously; the reaching arm has less competition.
After: the hip and thigh read more continuously; the reaching arm has less competition.
Before — clothed mass detail.
Before — clothed mass detail.
After — bands quieted by native deposits and lifts. Residual texture remains.
After — bands quieted by native deposits and lifts. Residual texture remains.

The head and neck construction is pixel-exact across the retained round. This is a tonal revision, not another pose edit or a new generated reference.

The critic changes the next proposal

Mass

Retain: quieter clothing, continuous compression.

Elliptical opening

Reject: a bright patch becomes a false highlight.

Contour omission

Reject: absence reads as a missing body fragment.

Mass branch: 571 new commands. Retained as the basis for the final accents.
Mass branch: 571 new commands. Retained as the basis for the final accents.
Elliptical opening: 676 new commands. Rejected despite successful numerical execution.
Elliptical opening: 676 new commands. Rejected despite successful numerical execution.
Developed after the opening failed: 700 commands crossing the silhouette. Rejected because the missing form becomes the subject.
Developed after the opening failed: 700 commands crossing the silhouette. Rejected because the missing form becomes the subject.
Return to the mass branch; nine authored marks reinforce the arm, thigh compression and floor.
Return to the mass branch; nine authored marks reinforce the arm, thigh compression and floor.

The two rejected branches are not merged into the final image. Numerical progress toward a tonal intention did not override the visual judgment.

What actually happened

  1. Restore the exact native scalar canvas from the prior round.
  2. Author tonal intentions from that canvas: group garment values, test an elliptical lift, then test an opening crossing the outer contour.
  3. For each branch, propose native deposits and lifts and judge their actual raster effect against the declared field. All three begin from the same source.
  4. Compare the complete native drawings. Keep the mass branch, reject the two openings.
  5. Restate a few arm, compression and ground passages with nine authored native commands. Replay the entire selected history from blank paper.

The masks and gestures are assistant-authored. The code does not understand anatomy or choose an artistic direction. The tonal fields guide execution; they are not pasted into the final canvas.

Mass intention · Opening intention · Omission intention · Mass scope · Omission scope ·

Critique and stopping

Useful gain: the clothes are less assertive as a repeated pattern, and the body’s compression is easier to read as one mass. The arm and ground accents make a modest contribution.

Still unresolved: the softened garment risks becoming a generic modeled volume. The hands remain simplified. Most of the composition is inherited. Neither omission trial found a convincing alternative.

Δ remains open. Keep this working revision for comparison. Do not declare artistic Hold because the numerical passages stopped. The next substantial artistic decision concerns where the drawing should remain definite and where it can genuinely fall away; these two erasure trials did not solve that.

Checks and evidence

  • All three tonal branches replay exactly and change zero pixels outside the declared garment rectangle.
  • The nine final accents have their own recorded scope; they deliberately extend to arm and ground.
  • The selected complete history replays exactly from blank paper: 3,896 commands.
  • Head/neck region: zero changed pixels in the selected result.
  • No new image generation, body fitting or reference download was used.
  • Visual judgment is by the main assistant, not an independent critic. No quality or emotion score is asserted.

· · · · · · Previous round