ENTRY 003
A Renderer That Reads Genetics
The renderer has been rebuilt. It no longer takes a list of appearance settings: it takes a gecko’s genes and works out what the animal should look like, using the same crested gecko genetics database that powers ReptiDex.
That changes what the site can be. There is now a morph builder, where you choose genes and watch the gecko redraw, and a morph library with a page for every morph the renderer can draw. Eighteen are covered so far: the three base colours, Hypo, Phantom, Axanthic and Fire, then Harlequin, Flame, Pinstripe, Quadstripe, Dalmatian, Empty Back, Tiger, Lilly White and Patternless, along with the pattern colours that make a Tri-Color or a Halloween.
The art it draws on was redrawn too. The gecko is built from real illustrated pieces, with guides for the pin rows, the belly line, the chin, and now the fringe and kneecaps, so a marking lands where it belongs on the animal rather than where a rule of thumb puts it.
Most of this month went into getting the genetics right rather than the pixels. Pinstripe turns out to be plain dominant, and the broken look people call a partial pinstripe is Tiger cutting across the pin rows. Quadstripe inherits separately, so the game gives it its own gene. Dalmatian decides whether a gecko has spots while a separate value decides how many, because a Super Dalmatian is a lot of spots rather than a different genotype. Phantom and Empty Back turn out to work the same way, as overlays over the pattern, one light and one dark.
Where the research is unsettled, the game makes a choice and says so. A new guide on the site sets out where the genetics come from and exactly where we simplify or invent, so nobody has to guess which parts are real.
Every gecko on the site is drawn from genes by this renderer. Nothing is concept art, and nothing is a finished sprite pulled from a folder. Change one gene and the animal changes; change the seed and the same genes give you a different individual, which is the point: a morph is a range, not a picture.
Next is pairing: choosing two animals and seeing what could hatch, using the same genetics. After that, the rest of the database, morph by morph.
Expected hatch date: early 2028.

