design.vecreal.com / workshop / V valley · round 3
Brand reference · workshop · round 3 · 2026-05-12
V valley · round 3 — R4-Bv (slight valley + widened legs) + V-height analysis
Letter-spacing -0.04em + Bold 700 locked in rev-10 (just propagated across consumers). R4-D abandoned — its "pull-inner-top-outward to keep slope parallel" approach made legs too skinny. New R4-Bv variant addresses the same valley-extension goal by pulling inner top INWARD (widening leg stroke top) while only lowering the valley a few pixels. Also analysis on V-height (0.66em vs 0.69em) embedded below.
What's in this round
R4-Bv — valley lowered 1 unit (y=32 → y=33) + inner top moved INWARD
from x=22 to x=23 (mirror on right: x=42 → x=41).
Result: leg stroke widens from 12 to 13 at top, tapers mildly to 11.1 at valley. Stays
substantial throughout.
R4-Bv+ — same widened legs, valley pulled to y=34 (2-unit drop).
Stroke: 13 top → 10.6 valley. More taper, more valley extension. For comparison vs R4-Bv.
Why "pull inward" not "pull outward": R4-D's approach was to keep inner slope parallel to outer slope by pulling inner top OUTWARD (toward outer) when the valley dropped. That kept legs uniform but absolutely thin (9.43 stroke width vs original 12). R4-Bv's approach is to pull inner top INWARD (toward center) so the leg stroke widens at the top, compensating for the natural taper that comes from a deeper valley. Trade-off: legs taper slightly instead of staying parallel — but they stay full-bodied.
V-height axis — 0.66em (locked prior) vs 0.69em (proposed). See analysis below the spec table. Both heights shown across variants for visual confirmation.
Section A · Locked rev-10 system reference — Bold 700 + tracking -0.04em + dot 0.23em
Single row, R4-A at 80 px to confirm what's locked. This is the system the V geometry renders against. Everything below uses this state.
Section B · V valley variants — at locked 0.66em
All five variants at 80 px, V height 0.66em. Look at the valley region — R4-Bv and R4-Bv+ show the operator-proposed "lower + widen" geometry that addresses the R4-D skinny-leg problem.
Section C · V valley variants — at proposed 0.69em
Same five variants, V height bumped to 0.69em. ~4.5% taller V relative to "ecreal".
Section D · V geometry isolated — large, no markers
Strip text + dot to focus on the V alone. Inside-vertex treatment and leg stroke read cleanest here.
Section E · Close inspection — top picks at 140 px, V height 0.66em
R4-A, R4-B, R4-Bv side-by-side at display size. The R4-Bv valley extension and leg widening reads most clearly at this scale.
R4-A ⇄ R4-B ⇄ R4-Bv at 140 px
Section F · V-height direct comparison — R4-Bv at 0.66 vs 0.69, 140 px
Same variant, only height differs. ~4.5% taller V, ~5 px at 140 px display.
R4-Bv · 0.66em ⇄ 0.69em at 140 px
Section G · Small-size readability — 24 / 32 / 48 px, both heights
Valley refinement should not break favicon / sidebar / breadcrumb scale.
| 24 px nav / sidebar | 32 px breadcrumb | 48 px body | |
|---|---|---|---|
| R4-A · 0.66em | ecreal. | ecreal. | ecreal. |
| R4-B · 0.66em | ecreal. | ecreal. | ecreal. |
| R4-Bv · 0.66em ⭐ | ecreal. | ecreal. | ecreal. |
| R4-Bv+ · 0.66em | ecreal. | ecreal. | ecreal. |
| R4-Bv · 0.69em | ecreal. | ecreal. | ecreal. |
Spec summary
| Variant | Valley | Inner top (left, right) | Leg stroke (top → valley) | Read |
|---|---|---|---|---|
| R1 | sharp · y=32 | 22, 42 | 12 → 11.7 | sharp baseline · all-vertices cohesion |
| R4-A | sharp · y=32 | 22, 42 | 12 → 11.7 | current · rounded outside, sharp inside |
| R4-B | rounded · y=32 (dips ~31.3) | 22, 42 | 12 → 11.7 | cohesion-fix · all 3 vertices rounded |
| R4-Bv ⭐ | rounded · y=33 (dips ~32.3) | 23, 41 (widened) | 13 → 11.1 | mild valley extension + widened top · NEW |
| R4-Bv+ | rounded · y=34 (dips ~33.3) | 23, 41 (widened) | 13 → 10.6 | more valley + same widening · NEW |
V-height analysis · 0.66em vs 0.69em
The numbers. 0.66em vs 0.69em is a 4.5% relative jump. At 80 px wordmark display: V is 52.8 px (0.66) vs 55.2 px (0.69) — ~2.4 px taller. At 140 px: 92.4 px vs 96.6 px — ~4.2 px taller. At small scale (16/24/32 px): difference is sub-pixel and disappears in rendering.
Relationship to cap-height. Inter Bold cap-height is roughly 0.72em. The V at 0.66em sits below caps (clearly its own glyph; reads as a chevron mark beside lowercase). At 0.69em the V approaches cap-height — closer to "what 'V' would look like if Vecreal were typed in mixed case." Past 0.72em the V starts to feel like a true cap.
Mass distribution with the just-locked Bold system. rev-10 just bumped "ecreal" from Semibold 600 to Bold 700 and eased tracking from -0.05em to -0.04em. The locked V height of 0.66em was set during the Semibold era when "ecreal" was visually lighter relative to the V. With Bold "ecreal," the text now carries enough mass to hold its own against the V — meaning the V at 0.66em is now better balanced than it was before the weight bump. The system rebalanced toward the V/text equilibrium without changing the V's height.
What 0.69em would do. It would push V back to dominance relative to "ecreal." That's a fine move if the goal is a more "mark + wordmark" feel (the V reads as a separate symbol next to text). It's the wrong move if the goal is "one cohesive word with a specially-styled first letter" (V integrates with "ecreal"). Vecreal's positioning is institutional — both reads can be institutional. But the just-locked Bold already pulled the system toward institutional weight. Adding V height on top is over-correction.
Recommendation: hold 0.66em. The system is already rebalancing toward institutional via Bold text. The V at 0.66em with Bold "ecreal" reads equilibrium-V-text — neither side dominating. 0.69em would tip it back to V-dominant. Save 0.69em as a future-revisit knob if the next 6 months of brand surfaces (decks, social, UI titlebars) suggest the V needs more weight relative to text. Right now: held.
If you disagree — the case for 0.69em is that Bold "ecreal" creates so much horizontal mass that even a slightly taller V is needed to keep the V from feeling undersized. The Section C row at 0.69em + Bold should read against Section B at 0.66em + Bold to verify. I'd watch for "V looks shrunken" (argues for 0.69em) vs "V/text feel cohesive" (argues for 0.66em).
Lock candidates — pick V geometry + V height pair. My pair-pick:
R4-Bv + 0.66em.
Once you call it, I'll lock and propagate (favicon, monogram, LinkedIn assets, landing) atomically.
Dot at 0.23em stays — the lighter, slightly-extended valley in R4-Bv doesn't change the dot/V
ratio enough to need a dot revisit.
Vecreal · brand workshop · 2026-05-12 · v-valley round 3