Dave LeCompte (he/him)'s Avatar

Dave LeCompte (he/him)

@bigdicedave.bsky.social

Game Developer, 3d printerer, dancer, movie-goer, theater-goer, road tripper, loser of found things. https://bigdicedave.itch.io/

20 Followers  |  64 Following  |  61 Posts  |  Joined: 15.11.2024  |  2.2953

Latest posts by bigdicedave.bsky.social on Bluesky

beginning drawing the second black quad on the second draft

beginning drawing the second black quad on the second draft

red and black quads complete, ready for the new blue SVG

red and black quads complete, ready for the new blue SVG

missing blue quad from before now being drawn

missing blue quad from before now being drawn

I've built a greedy algorithm that tries (not very hard) to reduce the penup travel times by picking a next path starting point close to the last path ending point. This leads to hard to predict fill patterns.

I've built a greedy algorithm that tries (not very hard) to reduce the penup travel times by picking a next path starting point close to the last path ending point. This leads to hard to predict fill patterns.

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Here you see two red quads and two blue quads. If you looked at the finished work, you know that there are intended to be three blue quads. If I wanted to lie to you, I could say that I'm going to draw the third quad next - but actually, this is an error, and I abandoned this print at about this point. My SVG generation generates one black, one blue, and one red SVG, and I had sent the third blue quad to a debug SVG which doesn't go to the plotter. I hadn't noticed this at generation time, only as the plotter had gone into action, so I reloaded the plotter, regenerated the blue SVG, and started over.

Here you see two red quads and two blue quads. If you looked at the finished work, you know that there are intended to be three blue quads. If I wanted to lie to you, I could say that I'm going to draw the third quad next - but actually, this is an error, and I abandoned this print at about this point. My SVG generation generates one black, one blue, and one red SVG, and I had sent the third blue quad to a debug SVG which doesn't go to the plotter. I hadn't noticed this at generation time, only as the plotter had gone into action, so I reloaded the plotter, regenerated the blue SVG, and started over.

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
beginning to plot the first red quad

beginning to plot the first red quad

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I generated full sized flow fields that I subsequently clipped to the polygons needed, so each of the blue polygons, each of the black polygons, and even each of the red polygons are unique.

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

For this one, I generated 7 different flow fields - the two red pieces shared a flow field, but each color had a different profile of "dispersion" or how wiggly it is, and blue and black had three different orientations. I could have just generated one tile's worth of flow field paths, but no.

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
a hexagon comprising 8 smaller half-hexagon quadrilaterals, in red, blue, and black ink. The quads are filled with flow fields, which look to me a little like wood grain. The red pieces are the smoothest, followed maybe by the blue, then by the black. I find it interesting that I can't tell for certain whether blue is wigglier than the black or vice versa.

a hexagon comprising 8 smaller half-hexagon quadrilaterals, in red, blue, and black ink. The quads are filled with flow fields, which look to me a little like wood grain. The red pieces are the smoothest, followed maybe by the blue, then by the black. I find it interesting that I can't tell for certain whether blue is wigglier than the black or vice versa.

Flow Field Half-Hexagon Quads

Still, largely experimenting with pen swaps and flow fields.

#iDraw #penplotter #flowfield #proceduralart

22.09.2025 04:09 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
beginning to draw red squares, this time in red, this time mostly vertical

beginning to draw red squares, this time in red, this time mostly vertical

I believe that the red flow field has a little more wiggle, a little more angular variance than the black field, just for extra flavor

I believe that the red flow field has a little more wiggle, a little more angular variance than the black field, just for extra flavor

almost done filling in the red squares

almost done filling in the red squares

then, when the black was finished, I took the black pen out of the plotter and inserted a red pen, drew a separate SVG for the red squares

21.09.2025 16:47 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
pen plotter drawing an 8x8 grid of squares

pen plotter drawing an 8x8 grid of squares

pen plotter filling alternating squares with flow fields, approximately horizontal, using a black pen

pen plotter filling alternating squares with flow fields, approximately horizontal, using a black pen

two rows of black squares completed, starting on the third row

two rows of black squares completed, starting on the third row

I plotted the black lines between the squares first, then the flowfields for the black squares

21.09.2025 16:47 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
pen plotter drawing with flow fields alternately red-vertical and black-horizontal, forming a chessboard. The finished drawing is 8 inches square, each square being an inch on a side.

pen plotter drawing with flow fields alternately red-vertical and black-horizontal, forming a chessboard. The finished drawing is 8 inches square, each square being an inch on a side.

Flow Field Chessboard

I took my flow field code and generated two flow fields - one mostly horizontal, one mostly vertical, then clipped the resulting paths into individual squares. I collected the squares into two SVGs, plotted them with different pens
#idraw #flowfield #procedural #penplotter

21.09.2025 16:47 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I also filled the gaps on the outside of the 25 ring with circles tangent to pairs of 25-beads and the containing circle.

I finished it all off with what I call a "circle fractal", recursively tangential circles filling in the corner spaces of the square.

19.09.2025 14:16 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

There were 9 gaps between the 9 beads, and only 4 of those gaps had been filled going "up" from 4, so I generated circles for all 9 gaps and checked for collisions with the 4 circles I had already generated, and adjusted the colliding circles.

19.09.2025 14:16 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Each ring is randomly rotated relative to each other, and to the square frame. This breaks up symmetries that would otherwise appear.

I then inserted more circles, nesting circles going "up" from the ring of 4 beads to the ring of 9, which gave 4 nested circles.

19.09.2025 14:16 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I started on the outside, with a square the size of the drawing space. I then nested 25 circles inside that, using sin(2pi/25) to find the spacing of the 25 "beads" - this gave me the radius of the ring of 25 beads, and how far to inset the next ring. I proceeded similarly with 16, 9, and 4 "beads".

19.09.2025 14:16 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
pen plotter drawing of concentric circles, made up of 25, 16, 9, 4, and 1 "beads". The finished work has a vaguely astrological feel.

pen plotter drawing of concentric circles, made up of 25, 16, 9, 4, and 1 "beads". The finished work has a vaguely astrological feel.

"circle fractal" of circles filling the space in the corner between the ring of 25 circles and the square. I started with the shapes of the left side and bottom side of the drawing, along with the outer circle, then found a circle tangent to all three shapes, which gave me three new roughly triangular regions which I could recursively fill. I proceeded until the circles had got down to 1/20 of an inch.

"circle fractal" of circles filling the space in the corner between the ring of 25 circles and the square. I started with the shapes of the left side and bottom side of the drawing, along with the outer circle, then found a circle tangent to all three shapes, which gave me three new roughly triangular regions which I could recursively fill. I proceeded until the circles had got down to 1/20 of an inch.

the beginnings of the ring of 25 beads being drawn

the beginnings of the ring of 25 beads being drawn

"Squared Circles" for 9/16/25
September 16 2025 was a "Pythagorean date", each element a square, the 2-digit year being the sum of the other two numbers. I nested a sequence of 4, 9, 16, 25 circles inside a square.
#idraw #penplotter #circles #pythagoras #procedural

19.09.2025 14:16 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I also added a greedy "travelling salesman" sorter that takes the paths as generated by the flow field code and attempts to reduce pen up travel time by reordering and reversing paths as necessary.

15.09.2025 15:46 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
"River" flow field, drawn in blue gel pen, flow lines are largely coherent, mostly going from top left to bottom right of the image.

"River" flow field, drawn in blue gel pen, flow lines are largely coherent, mostly going from top left to bottom right of the image.

Beginning of drawing the flow field, individual strands are being drawn, some spatial coherence is visible.

Beginning of drawing the flow field, individual strands are being drawn, some spatial coherence is visible.

Nearly finished drawing, a few gaps are visible

Nearly finished drawing, a few gaps are visible

only a few gaps remain, resembling onion skin

only a few gaps remain, resembling onion skin

New Flow Field

variation on a theme - constrained the flow field to a circle, constrained the vectors in the field between -35 and 75 degrees, for a more "river" flow visual.

#iDraw #penplotter #generativeart #proceduralart #flowfield

15.09.2025 15:46 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
debug output, with heading of the flow field lines encoded to HSV.

debug output, with heading of the flow field lines encoded to HSV.

I also experimented with outputting two SVGs as I implemented this - this is the "debug view", where I use color to indicate the heading of the paths, using a HSV colorspace.

15.09.2025 00:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I'm also experimenting with how high the pen should be in its "lifted" state - the iDraw uses a servo to press the pen into the paper. If you start the pen too low, you're putting unnecessary strain on the servo. For this drawing, the pen started at the thickness of two quarters stacked up. Seems OK

15.09.2025 00:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

From each point, I trace out a path using simplex noise, stopping when I reach the edge of the draw area, or bump into a path, including the path I'm currently generating.

Since the seed points are generated flooding out from a single center point, the fine detail of the plot also fills out.

15.09.2025 00:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

This work, like the rest of my works, are created in Python, generating a SVG file, which I feed to the plotter.

The Python script for this begins with a Bridson Blue Noise poisson disk point set.

15.09.2025 00:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Completed drawing, with space-filling curved blue lines following a "flow field", generated from simplex noise

Completed drawing, with space-filling curved blue lines following a "flow field", generated from simplex noise

Flow field drawing in process - the early drawing has some "tendrils" reaching out to the edge of the draw area, but most of the detail is within a circle - this is a result of the algorithm that generated the "seeds" that lead to the paths, which starts at a point and spirals out.

Flow field drawing in process - the early drawing has some "tendrils" reaching out to the edge of the draw area, but most of the detail is within a circle - this is a result of the algorithm that generated the "seeds" that lead to the paths, which starts at a point and spirals out.

Beginning of the drawing - some long tendrils are visible and many arching short paths connecting some of the tendrils, resembling contour lines from a mountainous region

Beginning of the drawing - some long tendrils are visible and many arching short paths connecting some of the tendrils, resembling contour lines from a mountainous region

Flow Field drawing

Flow field, where the direction of the flow comes from simplex noise. Plotted using blue EnerGel gel pen with an iDraw H SE plotter.

#iDraw #PenPlotter #ProceduralArt #GenerativeArt

15.09.2025 00:31 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
A poster with the words MATH JAM in large letters, and small numbers and letters filling the rest of the poster, drawn in blue ink.

A poster with the words MATH JAM in large letters, and small numbers and letters filling the rest of the poster, drawn in blue ink.

A poster for "Math Jam", a get together of recreational math folks mathsjam.com

using Hershey fonts, bridson blue noise, and line segment SDFs.

#iDraw #penplotter #math #mathjam #mathsjam

03.09.2025 03:11 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
a pen plotter drawing of a rectangle around a hexagon around a bunch of small circles, around a pair of medium sized circles, filled with a set of parallel lines. There's some extraneous lines in the upper right hand side that I need to sort out, I think my plotting bed is not quite level.

a pen plotter drawing of a rectangle around a hexagon around a bunch of small circles, around a pair of medium sized circles, filled with a set of parallel lines. There's some extraneous lines in the upper right hand side that I need to sort out, I think my plotting bed is not quite level.

#iDraw #penplotter #proceduralart
an exercise in clipping shapes (circles) against other shapes (circles, half-planes) specified as SDFs.

Reminds me a little of a guitar or a fried egg.

01.09.2025 01:50 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
apollonian gasket of 13 circles inside a larger circle

apollonian gasket of 13 circles inside a larger circle

close up of recursive tangent circle chains

close up of recursive tangent circle chains

circle inversion of the outer gasket

circle inversion of the outer gasket

Drawing an Apollonian Gasket with a circle inversion on an UUNA Tek iDraw H A3 pen plotter

#penplotter

22.08.2025 17:06 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

But seriously, one should probably TEST one's customer facing services, including the various channels on the "Contact Us" page of your website.

Customer feedback is gold. At the very least, have a human acknowledge it.

28.01.2025 00:17 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

That email immediately bounces back, with information useful to IT staff for debugging the problem, but basically saying that an email address must be approved ahead of time to send mail to the customer feedback address.

This is silly misconfiguration of the mail server, we've all done it.

28.01.2025 00:17 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Has this happened to you?

You have a less-than-perfect experience at a local business (e.g. a movie theater), you go to their website to inform them of what was less than ideal, provide a few suggestions of how to make it better next time, avoid the bad experience.

28.01.2025 00:17 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

This was my first exposure to full-on photogrammetry; building 3d environments from a selection of photos - more than just panorama stitching, the user was able to indicate the geometry of objects in the world, with enough effort.

22.01.2025 19:26 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I know there's 3 different tension holes feeding in to the machine, and maybe I need to get better about adjusting on the fly.

Worst case, I rip the entire hat apart, start over from scratch. I kind of want that hat to be ready by March, for the color scheme to be seasonally relevant.

21.01.2025 17:14 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

For this hat, the dark green felt "tight", and I noticed that some loops didn't come off the hooks correctly, sometimes stuff got snagged and the yarn split. I have ripped out all of the dark green, and will make my first attempt of re-hanging the hat on my machine.

21.01.2025 17:14 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

@bigdicedave is following 20 prominent accounts