One SREF, different subjects: what actually carries across?

Observation is different from an experiment

The archivist and miniature bridge shown below use the same style reference. They offer a useful look at recurring cool–warm relationships, soft surfaces and fine boundaries. Their scene and format differ, so they do not establish a controlled causal comparison.

Use the procedure below to run and record your own comparison. The examples illustrate visual traits rather than the causal effect of a parameter.

Prepare three readable subjects

Try a seated person, a ceramic cup and a small bridge. Describe each main subject and only the necessary setting. Avoid giving one elaborate lighting instructions while leaving the others neutral.

1. Keep the model, reference version, SREF, aspect ratio, style weight and stylize constant.
2. Retain several results for each subject, not only the one closest to your expectation. Record settings and date.
3. Compare major color areas before edges, surface and spatial hierarchy.
4. For a second round, change one setting only. Keep its records separate; randomness can still produce variation.

Ask the same questions of each result

Palette: are warm and cool colors used in comparable roles? Edges: are boundaries crisp, drawn or diffused? Surface: is paper, grain or glaze still visible? Space: how is the subject separated from its surroundings?

When a quality appears in only one subject, label it as possibly subject-dependent. Resist turning an observation from one attractive image into a universal property of a reference.

Write a bounded conclusion

A useful note might say that blue-green shadows remained evident across these three subjects, while facial rendering varied. It should not promise identical color in every future image.

When sharing a comparison, state the date, versions and whether the settings were held constant. Selected finished works are valuable examples too, as long as the reader can tell what kind of evidence they provide.