The marking standard

How verstag determines the label's size, margin and contrast for each image.

In „Automatisch“ (Automatic) mode you do not set the label size; legibility does. verstag computes size and margin per image from a versioned rule set — the marking standard.

What counts is the largest text belonging to the marking: „AI“ has to stay legible, the qualifying word may recede. That holds for the platform wordmark and for the EU capsules alike.

The explainer for it is headed „So wird die Labelgröße berechnet“ (How the label size is computed) and is reachable from two places: the calibration dialog under Settings → Labels, and the defaults form, there only in automatic mode. In the editor a short version sits behind the info icon.

The signed-in product speaks German, so its labels are quoted in German throughout, with the English rendering in italics.

„Automatisch“ and „Manuell“

„Automatisch“ is the default. In that mode the two sliders for size and margin disappear; a line with the result stands in their place. The stored numbers stay behind — they are the fallback for manual mode, not deleted.

Free placement is not offered in automatic mode. If you had chosen a free position and switch to „Automatisch“, it is reset to the default anchor.

Conversely, „Manuell“ adopts the numbers just computed as its starting point. The sliders sit where the standard left them.

Even in manual mode the margin slider only appears for an anchored label — under a freely dragged one it controls nothing.

The margin

The margin is measured to the label's visible edge, not to the file's: the artwork's transparent margin on the sides it rests against is subtracted, in a way that leaves each of those sides at least the required distance.

Rotation does not change the size. It only affects the margin, because it decides which edge of the artwork rests against which image edge.

The template's aspect ratio

How wide the label gets at a given height comes from the SVG file. For the shipped templates the value comes from the viewBox. Where a template of your own states neither dimensions nor a readable viewBox, an assumption applies — calibrate an uploaded variant and look at the result.

When the image is too small

Where the visible type no longer fits the output size, nothing is burnt in — and the output is not produced at all. On screen it then reads: „Dieses Bild ist zu klein für eine normgerechte sichtbare Kennzeichnung.“ (This image is too small for a standard-conforming visible marking.)

The way out is to switch size and margin to „Manuell“; the button beside the warning does exactly that.

Uncalibrated templates

A template that has not been calibrated falls back to the stored manual numbers. The label is still burnt in — it is simply not computed. „Unkalibriert“ (uncalibrated) may stay a permanent state; the editor states it as a hint and uncalibrated variants are filtered out of no picker.

If the output size is not known yet — no format chosen, video metadata still loading — the interface says nothing about it.

What gets saved records which version of the standard produced the result. For an uncalibrated variant, or an output with no known dimensions, that entry is omitted; the size stays yours.

Contrast

Contrast has a minimum ratio of 4.5 : 1. Both variants of a family are measured against the background where the label sits; for a variant its worst case is what counts. The result appears in the editor in „Automatisch“ mode as the contrast line — verdict, value, minimum, and the name of the variant that will be burnt in.