Each work is a pair. The left panel is a photograph as it was made. The right panel is the same photograph carrying one intervention derived from its own measurements. Side by side, the difference often lives at the threshold of vision, and it is stated exactly, down to fractions of a single tonal step. The numbers are the work, and anyone may measure them again. You are offered almost nothing to see, and everything to check. · · · The short-wavelength channel has been defocused the way the eye's own optics would, so the blue is now as sharp as the modeled eye could ever have received it in that room. The eye's optics carry no chromatic correction. In the standard accounts of the aberration, blue focuses roughly 1.3 dioptres in front of yellow-green, which means that in a dark gallery, accommodated on the figure and the warm wall, a saturated blue field physically cannot come to focus on the retina. It arrives as a disc whose angular diameter equals pupil diameter times defocus. 6 mm × 1.3 D = 0.447°, which at this lens's field of view is 6.9 pixels. The camera, whatever its own aberrations, rendered that blue rectangle sharper than the model says any eye in the room could have received it. The photograph has been given a model of the eye's optics. Converted to LMS cone space, the S channel alone was blurred with a 6.9-pixel disc, a disc rather than a Gaussian because geometric defocus is a disc, and L and M were left untouched. The luminance change is 0.00000 linear on average and 0.00538 at the single worst pixel, and this required no tuning. S contributes almost nothing to luminance, so an operation confined to it changes color while leaving brightness all but untouched. Where it landed, measured in the S channel, the blue field's lower edge softened 32%, the figure's silhouette 27%, and the left edge 21%. The largest single change, 38 levels, sits at the blue field's lower-left corner, the hardest chromatic edge in the picture, exactly where the modeled eye fails. Presentation. The mount is neutral 57. Neutral because the work lives entirely in the S channel, where a tinted surround would pre-adapt the very cones the work addresses. Dark because the stated blur assumes a 6 mm dark-adapted pupil. And exactly 57 because that is the blue field's own luminance where it meets the mount. The step across that join measures +0.00032 linear for both panels, so along the entire right edge blue meets gray in almost purely chromatic contrast, with next to no luminance step, a border the luminance motion system has almost nothing to hold. To check. 213,406 pixels move, none by more than 38 of 255. Download this work's original file, 08_1.3_dioptres.png, not a screenshot or a marketplace preview, attach it to an AI that can run code, and paste this: Run this exact line on the attached file, changing only the file path if needed. Show me its exact output, and say whether that output is exactly: peak 38 changed 213406 python3 -c "import numpy, PIL.Image as I; c=numpy.asarray(I.open('08_1.3_dioptres.png').convert('RGB')).astype(int); a=c[690:2991,690:2991][::3,::3]; b=c[690:2991,3336:5637][::3,::3]; d=abs(b-a); print('peak', d.max(), 'changed', (d.max(2)>0).sum())"