Drew Ink The year

x-height

The height of a lowercase x, which governs how large a face looks far more than its point size does.

x-height: a letter in the house face with the x-height marked in vermilion.
x-height, drawn, Drew Ink. Drawn 2026-09-10.
x-height: a letter in Drew Slab Regular with the x-height marked in vermilion.
x-height, drawn, in Drew Slab Regular, Drew Ink. Drawn 2026-09-29.

Near it

Cap heightAscenderOvershootText faceOptical sizeDutch type

Legibility and readabilityOld-style figuresSmall capsTittle

What it is made of

THE DESIGNED RATIO IS NOT ONE NUMBER: eight families move the x-height and fifteen keep itfifteen families keep the base drawing's x-height, 12 units of a 16-unit cap, 0.7500; eight move it in the family engine, from Drew Display at 0.6500 to Drew Micro at 0.8625, with Engrave 0.6875, Terrace 0.8187, Press 0.8250, Poster 0.8375, Gazette 0.8438 and Compact 0.8500read from the family engine's own settings (the xheight each family declares), 2026-09-24. CORRECTED: until then this row said all 23 share one grid with no variation, read from the families' face.json design blocks. Those blocks are copied from the base drawing and do not record the move, so the row read a block that does not describe eight of the faces
measured off the ink, which is the drawing plus the strokex-height over cap height runs 0.7100 on Drew Display to 0.9314 on Drew Poster, a spread of 1.312 to 1 across 23 families. In four of the eight families that move the x-height, Compact, Gazette, Micro and Press, the x's ink now sits on the designed line (Micro 604 of 1000 against a drawn 604); in the other four it still stands above it, Display by 42, Engrave by 36, Terrace by 49 and Poster by 66, and so does the x of Extended, Hairline and Round against the base line, while the u of all seven sits on it: the x's top diagonals are left uncut, a fault in the foundry, found 2026-09-24 while re-measuring this record and owed to itmeasured over all 23 house families off the INK rather than the table, on the fixed shelf, 2026-09-28. The 0.7271 to 0.9443 this row carried on 2026-09-24 was measured before the x-height cut was folded in; the 0.6990 to 1.0257 before that, on the shelf before the baseline fix was measured on the shelf before the baseline fix and does not survive it
large x-heightDrew Micro at 0.8629 on the ink and 0.8625 by the drawing, cut for small sizes, where a large lowercase is what keeps a letter legible; on the ink alone Drew Poster's x stands higher, at 0.9314reads bigger at the same point size, and shortens the ascenders
small x-heightDrew Display at 0.7100 on the ink and 0.6500 by the drawingmore classical, more air, needs a larger setting
the table and the ink are not the same numberOS/2 sxHeight and the top of the drawn x agree on all 23 families at Regular, and differ on 3 of the 84 files, by 1 unit in 1000: Extrude Light and Bold, and Stencil Semiboldnot a design difference, and still not the same number: this row says the ink, because sxHeight is a value somebody typed and the outline is where the curves go