Ensō

The work and its context

Ensō places a single dark circular gesture on a pale, finely textured field. The gesture is simple enough to recognise at once, but it does not remain uniform as the eye follows it. A substantial black arc gives way to grey strands, and those strands can almost disappear before reaching the stroke's beginning. The empty centre is therefore enclosed with different degrees of certainty. In the four examples examined here, density and interruption give each circle its own balance.¹²³⁴

Matto released the 1,000-work series through Art Blocks Factory in March 2021. The artist explicitly identifies sumi-e, or ink wash painting, as its inspiration and describes the program as an attempt to model the making of an ensō. The platform's description names paper, brush, bristle-level ink quantities and ink flow among its variables. Those material references establish the project's premise; inspection of the deployed script shows how selectively they are translated into a drawing procedure.⁵⁶

The Matto dossier provides a wider context for the practice. Here the central question is narrower: how does a procedure produce the appearance of a gesture whose interest depends on variation within one stroke? The answer is visible in the difference between a ring's overall contour and the small changes of darkness along it. The work invites the viewer to move between those scales rather than treating the circle as a finished geometric sign.¹³

Conception and development

On his project page, Matto describes art and programming as long-standing pursuits. Martial arts and Eastern spirituality entered his life during college, while sumi-e became a meditative practice. He presents Ensō as a meeting of interests that had previously remained separate. Attempts to make tokenised work led through learning and experimentation with several languages before Ensō and Zenerative emerged. This is the artist's retrospective account, not a dated record of every development stage.⁷

The same page documents separate water paintings and physical prints associated with the project. Those extensions matter because they place the digital series alongside actual material acts. They do not establish that the generated circles are scans of handmade paintings. The inspected program constructs paper texture and brush marks directly; the resemblance to ink results from choices about points, opacity and depletion.⁷⁶

There is a productive tension in that construction. A circle can be specified with a compact mathematical rule, while a brushstroke appears to carry the history of a hand moving through time. Ensō gives that mathematical path an uneven supply of simulated ink. The resulting image preserves an apparent sequence even though the viewer normally encounters the completed mark. Its temporal suggestion belongs to the distribution of darkness, not to a continuously replayed animation of a brush.⁶²³

How the work was constructed

The complete 7,392-character indexed script was inspected as text, and four canonical renderers were examined directly. The project uses p5.js 1.0.0. Its main drawing is generated once within the draw callback, after which the callback continues to handle the signature state. The presence of a continuing callback should therefore not be mistaken for an endlessly changing circular composition. In the inspected renderers, the main mark remains stable.⁵⁶¹

Two mechanisms derive choices from the token hash. The program divides the hash into bytes and consumes those values sequentially for bounded selections. It also derives a numerical seed from a later portion of the hash and applies a sequence of bitwise transformations for further variation. The distinction matters: coarse properties such as a brush parameter are selected separately from the many smaller decisions that make the surface irregular.⁶

The canvas is square, sized to the smaller window dimension. The paper begins with a pale ground, usually with a slight hue, and receives 2,400 transparent curved marks. Their positions, thicknesses and opacity vary. These overlapping curves produce the fine crossing texture visible around and inside the circle. They are not fibres individually modelled as physical objects, but they give the surrounding field enough variation to participate in the image.⁶¹⁴

The brush is represented by a central point and successive rings of bristle positions. Ring count and density affect the number and spread of those positions. Each bristle then receives its own sequence of ink quantities along the stroke. The initial quantity varies and depends partly on distance from the brush's centre. The arrangement provides a broad mark with internal differences, rather than one opaque outline whose thickness changes uniformly.⁶

Ink usually diminishes as the sequence advances. A small conditional branch can allow a depleted bristle to regain ink once before fading again; another state sets it to zero. These rules create irregular persistence within the brush. They should be described as a stylised depletion procedure, not as a complete simulation of fluid flow, absorption or the pressure of a human hand. No such physical system was found in the inspected calculation.⁶

The path contains 1,600 positions around a circle, modified by independent stretch factors, orientation and direction. At each position the program draws points offset by the bristle layout. Their opacity follows the stored ink quantity, while point weight gradually decreases along the path. The combination makes one end comparatively forceful and the return more attenuated. The underlying route can continue around the circle even where the ink becomes too faint to make it readily visible.⁶

This separation between route and deposit is a key aesthetic decision. An opening does not necessarily mean that the mathematical path stopped. It can emerge because the marks along that path lost density. The viewer sees an interrupted gesture, while the procedure has followed a continuous organising shape. That difference helps explain why some examples seem complete as circles yet incomplete as strokes.⁶²³

The lower-right signature is drawn from lines and a small circle. Clicking toggles its display. In edition #0, one click removed the coloured stamp and a second restored it while leaving the main image unchanged. That observed behaviour supports the artist's description of the interaction. It does not make the work an animated painting. Reproduction still depends on the token input, script, library and initial display dimensions; the inspection did not test equivalence across every browser or screen size.⁶⁷¹

The mint

The collection page gives 22 March 2021 as the release date. The index records the first token on 21 March at 02:55:42, activation on 22 March at 02:11:50, and the second mint and scheduled start at 15:01:47 that day. Completion is recorded on 26 July at 10:58:59. These are distinct recorded events. In particular, the March release and July completion should not be compressed into a claim that the edition sold out immediately.⁵

The observed index reports 1,000 invocations against a maximum of 1,000 and a price setting of 0.108 ETH. Those values establish the current indexed supply and configuration. They do not independently establish the originally announced maximum, the opening public price or what each recipient paid. Historical changes to the configuration and the division between artist, reserved and public mints remain unresolved.⁵

The artist's account of related gifts supplies useful context about the project's surrounding activity, but it does not replace transaction evidence. This essay therefore does not calculate an exact public-sale duration or classify the first token as a public purchase. Such a reconstruction would need launch-era records and transaction-level analysis, with the meaning of each timestamp kept separate.⁷⁵¹

Copyright and licensing

The project record reports CC BY-NC 4.0 for the artwork and supplies the corresponding Creative Commons terms link. The deed allows sharing and adaptation with attribution, indication of changes and a noncommercial condition. This is the observed artwork label. A complete history of any earlier terms has not been reconstructed.⁵⁸

The p5.js 1.0.0 dependency has an LGPL version 2.1 licence in its tagged repository. That library licence does not itself grant permission to reuse Matto's program. No separate artist-code grant was located in the inspected sources. The artwork, artist script and supporting library are consequently recorded as separate rights questions. AB5D's CC0 dedication applies to the original essay prose.⁹⁶

Close reading of four outputs

The comparison uses editions 0, 1, 10 and 20, selected in advance as an early-edition group. Their token identifiers are 34000000, 34000001, 34000010 and 34000020 on Ethereum contract 0xa7d8d9ef8d8ce8992df33d8b8cf4aebabd5bd270. All four canonical renderers were inspected. The sample supports a comparison of particular outputs, not a statistical description of all 1,000 works.¹²³⁴

In #0, the circle is a slightly tilted oval. A dense black arc occupies the left and upper portions, while the right side and lower return separate into finer grey strands. Near the lower-left junction, a blunt dark end meets that attenuated return. The nearly closed shape makes the inequality of the stroke especially clear. The pale cream texture continues through its centre, and the yellow-green stamp supplies a small coloured counterweight near the lower-right corner.¹

In #1, a smaller circle leaves more surrounding paper. Its strongest black mark curves through the lower portion and ends prominently at the right. Moving around the left and upper arc, the stroke becomes increasingly faint until an opening at upper right is difficult to distinguish from a trace. The green-tinted field and green stamp make the darkness feel less isolated than in #0. The circle is legible as a whole even where its visible deposit almost vanishes.²

In #10, the heavy beginning sits high on the right, giving the image a different directional pressure. A broad dark arc descends along that side, becomes grey through the bottom and fades toward the upper-left endpoint. The gap across the top remains conspicuous. Fine strands at the lighter end encourage close looking after the overall shape has been recognised. A blue stamp and cool, faintly lilac field further distinguish this more open composition.³

In #20, a larger, almost closed circle occupies more of the available square. The left and top remain assertively dark; the right narrows and the bottom returns as a thin grey band to a rounded dark end at lower left. Compared with #1, the form feels more enclosing and less tentative. Its magenta stamp is a sharply bounded accent against the pale lavender field, whereas the main stroke's edges remain softened by overlapping marks.⁴

Significance and limitations

Ensō's achievement is to make a small vocabulary of circular path, irregular deposit and textured ground sustain careful comparison. Its reference to painting does not depend on disguising every aspect of its computational construction. The regular path remains legible, while the bristles complicate it. Across the four inspected works, closure becomes a matter of degree: the eye completes a shape even as it notices the marks falling away.⁶¹²³⁴

The signature interaction gives the viewer one modest intervention into that balance. Removing the stamp changes the relation between the central gesture and its surrounding field without authoring a new stroke. The work thus holds most of its decisions fixed while allowing a small change in presentation. Its strongest argument remains in the image itself: one constrained route can carry visibly different histories of apparent ink and exhaustion.⁶¹

Unresolved research questions:

• Recover dated launch-price and maximum-edition announcements and any later configuration changes.
• Separate reserved, artist and public mint transactions.
• Locate any distinct artist-code licence and establish the history of artwork terms.
• Compare the indexed program against a pinned chain retrieval and test display-size consistency.
• Broaden close reading beyond four early editions without treating the sample as a rarity survey.