marco/IsoMmo

Watch

Fork 0

Yellow seam where marsh meets dry savanna (dual-grid base bleeds through) #200

New issue

Closed

opened 2026-07-24 21:38:46 +02:00 by panda · 1 comment

panda commented

2026-07-24 21:38:46 +02:00

Collaborator

Reported from the client: at the marsh / dry-savanna border a thin yellow line shows under the marsh.

Diagnosis

BiomeTransition.Order ranks marsh last, i.e. highest priority, and dry_grassland fourth. The dual-grid draw (GridRenderer.DrawDualTile) therefore takes the lowest-priority corner as the base full tile and paints each higher biome over it through a corner mask (DualGrid.MaskAtOrAbove). On a marsh/savanna edge that means:

base = dry savanna, drawn as a full diamond,
marsh painted over it with a 16-way corner mask.

Wherever the mask's coverage falls a pixel short of the diamond edge, the bright yellow base shows through beneath the dark marsh. The pairing is the worst case in the palette: savanna is the lightest ground, marsh the darkest, so any gap reads as a hard yellow line. (The masks already carry a ~2px expansion added when the original 1px seams were fixed in #162 — evidently not enough for this pair.)

Things to try

Widen the mask diamond expansion further and re-generate the 16 masks (scripts/genmasks.py), then re-bake.
Or draw the dominant corner biome as the base when three or four corners share it, so the mask only has to cover a minority edge.
Or pre-multiply/clamp the mask edge so its outermost ring is fully opaque rather than feathered.

Definition of Done (delta)

No light seam is visible along a marsh/dry-savanna border at 1:1 zoom, checked in-client on a fresh DB.
The other biome pairings are re-checked for regressions (the fix must not re-introduce the 1px seams of #162).

Related: #162 (dual-grid borders), #137.

Reported from the client: at the **marsh / dry-savanna** border a thin **yellow line shows under the marsh**. ## Diagnosis `BiomeTransition.Order` ranks `marsh` **last**, i.e. highest priority, and `dry_grassland` fourth. The dual-grid draw (`GridRenderer.DrawDualTile`) therefore takes the *lowest*-priority corner as the **base full tile** and paints each higher biome over it through a corner mask (`DualGrid.MaskAtOrAbove`). On a marsh/savanna edge that means: - base = **dry savanna**, drawn as a full diamond, - marsh painted over it with a 16-way corner mask. Wherever the mask's coverage falls a pixel short of the diamond edge, the bright yellow base shows through beneath the dark marsh. The pairing is the worst case in the palette: savanna is the lightest ground, marsh the darkest, so any gap reads as a hard yellow line. (The masks already carry a ~2px expansion added when the original 1px seams were fixed in #162 — evidently not enough for this pair.) ## Things to try - Widen the mask diamond expansion further and re-generate the 16 masks (`scripts/genmasks.py`), then re-bake. - Or draw the *dominant* corner biome as the base when three or four corners share it, so the mask only has to cover a minority edge. - Or pre-multiply/clamp the mask edge so its outermost ring is fully opaque rather than feathered. ## Definition of Done (delta) - No light seam is visible along a marsh/dry-savanna border at 1:1 zoom, checked in-client on a fresh DB. - The other biome pairings are re-checked for regressions (the fix must not re-introduce the 1px seams of #162). Related: #162 (dual-grid borders), #137.

variant_banding_fixed.png

905 KiB

seam_fixed.png

410 KiB

panda referenced this issue from a commit

2026-07-24 22:13:20 +02:00

fix(client): remove biome-border seams (binary masks + point sampling); tone-match ground variants (#200)

panda referenced this issue

2026-07-24 22:30:04 +02:00

fix(client): remove biome-border seams and in-biome variant banding (#200, #201) #202

marco referenced this issue from a commit

2026-07-25 18:33:15 +02:00

fix(client): remove biome-border seams and in-biome variant banding (#200, #201) (#202)

marco commented

2026-07-25 19:06:15 +02:00

Owner

Risolta da PR #202 (dc0b132): il seam giallo marsh/savanna è stato eliminato rendendo binario il bordo delle mask dual-grid (scripts/genmasks.py, SOFT 0.18→0.015) e correggendo il dimensionamento dell'overlap in pixel di tile. Chiudo — completata.

Risolta da **PR #202** (`dc0b132`): il seam giallo marsh/savanna è stato eliminato rendendo binario il bordo delle mask dual-grid (`scripts/genmasks.py`, `SOFT` 0.18→0.015) e correggendo il dimensionamento dell'overlap in pixel di tile. Chiudo — completata.

marco closed this issue

2026-07-25 19:06:15 +02:00

No milestone

No project

No assignees

2 participants

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

marco/IsoMmo#200

No description provided.

Rows
Columns