Chronicle of black_son_of_gray's failures

[Smashter]

Whoa! Really impressive. Those are goergous. Thanks for sharing.

[ertyu]

i really like the third one.

this project reminds me of the marble machine

very cool

[jacob]

It reminds me of LOGO? (I might be dating myself here)

We want to see the machine that did this!

Might also inspire the build a CNC router project which seems to have stagnated somewhat.

[zbigi]

This brings me back to my childhood, I was writing simple programs on my C64 to generate moire patterns.

[black_son_of_gray]

We want to see the machine that did this!

So, to preface, my two biggest inspirations are these two guys:
https://robertbalke.de/about/
https://www.jamesnolangandy.com/
(There are others, but these are definitely the top)

Basically, I reverse-engineered Robert Balke's machine by staring at it really hard and crunching some numbers re: gearing. My turntable system is different, but the overall design is basically the same. I have made some minor modifications over time as I wanted to test different parameters, so it has slightly mutated from there.

Gandy's machine is just a thing of beauty in itself. If I was into machining, that is 100% the route I would go. The tubing drive he uses is genius, and he doesn't have the tolerance issues that crop up from using, ahem, wood as I have. If I went down this route, I would probably not come out of the workshop for a year, playing around with cams and needlessly trying to incorporate various CVT cone-type mechanisms.

Here's my hack job:



To give you some scale, the turntable is half of a 2'x4' mdf panel, so I can do some very large drawings. (Naturally, I can also do drawings as small as a couple inches).

And, a close-up of a gear-box:



You may be thinking to yourself, "wooden dowels aren't going to be great given the torque and so on". That is 100% correct; I had to swap them out for steel rod, which is way better. Otherwise, my only regret is that the brushes on those motors are damn near impossible to get at/replace without being destructive. I really hate that, because easily replaceable brushes doesn't seem like a tall design ask and basically would make the motors last forever.

The pen setup:


Using a syringe w/ ink instead:


For various technical reasons, I found it is best to have the syringe 'float' freely within a length of brass tubing that is clamped into the pen holder.

Personally, I think the liquid inks are more attractive (this is gold and oxblood):


Another pretty picture just for giggles:

Might also inspire the build a CNC router project which seems to have stagnated somewhat.

Yeah, there are certainly some parallels. Here is a screenshot of my simulation during a run:



At the end of each "frame" of the simulation, the location of the "pen" is stored. As long as the increment between frames is sufficiently small, the resulting string of essentially x-y points looks indistinguishable from a curved line and serves as a perfect set of coordinates to drive a CNC-type device that is holding a pen rather than e.g. a router.

Again, I'm not really innovating here. There are even turnkey solutions (e.g. https://www.axidraw.com/) for me to draw the results of a simulation! I have debated buying something like that, but haven't been able to justify it...

[suomalainen]

I don't think of it as a hack job at all. I love the "tinkering" aspect of it. Not to, like, give you a project or anything, but it would be kinda cool to see a video that consists of all the drawings you've done in series, each displayed for a few seconds with a caption that narrates what you did to tweak the machine in between or what you were going for, that kind of thing. To see the evolution and experimentation, so to speak.

[7Wannabe5]

What fun!

[black_son_of_gray]

Not to, like, give you a project or anything, but it would be kinda cool to see a video that consists of all the drawings you've done in series, each displayed for a few seconds with a caption that narrates what you did to tweak the machine in between or what you were going for, that kind of thing. To see the evolution and experimentation, so to speak.

I thought about this comment for a bit. First, about how I didn’t want to do a video (sorry, I’m not), but then second, about lessons that I’ve learned with this project. And I think there are very relevant ERE themes.

First, to kind of tangentially get at the tweaking process and to give you a teaser of the hundreds of images I have, I present to you three simulation runs:





They look pretty different, huh?

Well, they are...and yet they aren't. These three simulation outputs are from extremely similar simulation settings. In fact, all of the settings are the same in these simulations except that one rotor's rotation speed was very slightly changed relative to the other rotor (only two of the three rotors were actually spinning! Just imagine the variability from adding the third.). The top image has a ratio of 1.85:1, the middle has 1.90:1, the last has 1.95:1. That means the differences are in the range of ~5-6%. So one lesson here is that a tiny change in just one variable (there are maybe a dozen meaningful variables in the simulation) can dramatically change the results.

This is the kind of thing that happens all the time within a system. In this case, my system (the simulation) has a lot of interconnected variables that influence each other in different ways, and it turns out that small tweaks to certain variable have profound impacts, while other variables have relatively subtle or more predictable impacts - things like exaggerating already present features or changing the overall dimensions a little bit.

The parallels to ERE systems thinking are fairly obvious, and with a similar kind of tinkering element. We may think we know which of the variables in our systems are the subtle ones, and which ones are the wild ones, but it's really difficult to know that for sure until you tweak it a bit and see what happens. Sometimes those variables turn out to be the opposite. Sometimes seemingly really huge life events/changes (at the time) end up not really mattering that much. Sometimes a small change (maybe taking 5 minutes out of a day to focus on a specific, little thing), ends up making a huge difference in the trajectory of a life/relationship. This is yet another reason/argument for the statement: "it just takes time". It kind of sounds like woo, but it's also kind of ... true.

I think you used the correct words when you said "evolution and experimentation", because to my mind that is more to the core of what happens within the systems I care about and find more interesting. The term 'change' is usually used (e.g. climate change), but evolution kind of implies 'changes that themselves are changing' and I like that a lot more.

In principle, we might categorize systems into two kinds: static and evolving. Static ones can be 'solved' or eventually repeat or are otherwise relatively stable/predictable. There are interrelated variables, but not much complexity/emergence resulting from their interactions. There is a degree to which these systems are relatively narrowly limited, bounded, or constrained*. I would guess, given the ERE Wheaton Table's earlier levels being filled with the kind of concepts that have simple, calculable, 'answers' that are the same from one year to the next (hell, even a century later), that those progressing into the systems realm initially might think in terms of static systems. In some way, this is a continuation of the kind of thinking of earlier levels. It's not quite fully there, though, not fully capable of digesting/dealing with the chaos of time dependency and instability. However, life is an evolving system, even though on very short timescales (e.g. day to day, maybe even month to month), one can be fooled into thinking that it is static or simpler than it really is. I do wonder, therefore, whether there isn't a subdivision(s) hidden within 'systems thinking' that is worth distinguishing - in particular, because so many people seem to get stuck there.

*Indeed, the very straightforward earlier Wheaton Levels could also be described from the point of view of 'systems', just incredibly dimensionally reduced ones.

[7Wannabe5]

Brings to mind a short story by the brilliant Grace Paley in which much drama ensues, but one of the last lines, just as a bit of an aside, notes the small child engaged in his habit of snacking upon a few lead-sweetened paint chips. I believe it is in her collection, "The Little Disturbances of Man", but it might be found in "Enormous Changes at the Last Minute."

[mountainFrugal]

Very cool project bsog!

Have you played around with water soluble inks? You could potentially get some interesting happy accidents (ink/paper combos). As an example, doing a few different design layers in primary colors of water soluble inks and then do a layer of a different design with just water or another color of ink heavily diluted in water.

[black_son_of_gray]

Have you played around with water soluble inks? You could potentially get some interesting happy accidents (ink/paper combos). As an example, doing a few different design layers in primary colors of water soluble inks and then do a layer of a different design with just water or another color of ink heavily diluted in water.

Kind of?

Here are some examples of how I have experimented with media. I haven't spent much time deep diving in any particular direction, but I'd be really interested in hearing your thoughts about what you find exciting (if anything).

First off, paints:



Composition-wise, I think there is a lot of potential here. Oil paints might be ideal (long open time), but I'm not a huge fan of the harsh chemicals involved. IIRC, these were acrylics with a retarding medium mixed in. What I hoped for was this: A thick-consistency paint layer over a (dry) colored background (in this case mars black, but could be anything); the drawing machine was set up with a metal rod stylus instead of a pen; the stylus has different degrees of rounded off pointed tips; when the stylus swirls through the wet top layer of paint, it scrapes off the wet applied layer, revealing the color underneath; cool, colorful patterns are therefore 'painted'. The actual result from three test runs (above) gives me a little hope that this approach could be dialed in, but it'll take some work getting the paint consistency/stylus tip right, and I'm not exactly sure how I would avoid multiple colors getting muddy/over-blended (as you can kind of see in some of the tests). It's very abstract (duh), but it would be cool to mix somehow with figurative elements.

Paint and Ballpoint Pen

Excuse the poor lighting. Most of the inks that I have tried using my syringe method have been acrylic inks because of their ease of use, blending, and thin viscosity (close to water). But acrylic inks are... kinda just paint. So in this case, just to see what would happen, I painted a circular ring with the inks (dripped it on in a circle, then squashed it with another layer and rotated). Then, after that dried, I ran the drawing machine over top with a silver metallic pen. This was to test 1) whether the acrylic inks would warp the paper when it dried (the paper needs to lie flat on the machine to work well)- it did dry nice and flat; and 2) whether the pen would lay ink well over the dried acrylic, which can sometimes be 'slick' and give issues for a ball-point mechanism. It worked just fine!

Solar printing

This idea is kind of wild. For solar printing/cyanotyping, the substrate (paper, fabric, etc.) is coated in a UV-sensitive paint/dye and then exposed to the sun outside to set the image. To make the above picture, I drew a shape with my drawing machine on a sheet of clear plexiglass using a syringe and india ink. I then laid that plexiglass sheet over a prepped paper as an overlay. The india ink pattern blocked the UV (those parts stayed white). After exposure, I thoroughly washed the paper to remove the excess dye and let it dry. What's cool about this is that 1) I can now put my drawing on anything that I can apply the dye to and wash (so, e.g. fabrics), and 2) I can reprint the same design multiple times. These can also pretty easily be mixed with other overlays/negatives/printing methods to greatly expand composition opportunities.

It would be cool to play around with watercolor pens, or run a waterbrush over a watercolor-pencil drawing, etc., but I haven't gotten around to it.

Masking + Pen


I wanted to see how cleanly masking could be done. It turns out, basic scotch tape does a wonderful job. It removes cleanly, is thin enough for the pen tip to 'ride' over, and can be cut accurately with an X-acto knife. I was super happy with the Steller's Jay, so I feel pretty confident that almost any composition could be masked with decent results. I haven't seen anyone with a drawing machine explore this direction, so maybe that is the place to go?

[mountainFrugal]

I did some quick and dirty materials experiments for you with some scrap paper, personal tools and classroom tools. Not all of this will be completely relavent for your machine, but maybe they will all spark ideas.

I think starting with toned paper and then going darker (black extreme) or lighter (white extreme) always adds to compositions. You should play with toned paper! It is really fun to think in both directions of value by starting in the middle rather than at one extreme and working lighter or darker. The ink is Copic and the paint pens are labeled.

For permanent ink... Platinum black carbon ink is my personal favorite. It is actually black and does not smear too much when completely dry. You can dilute it with water before you apply to paper and do various layering.

The Copic Gasenfude brushpen (or similar) might offer you some more line variation. The example is from the same pen.Copic pens are also the highest quality. You can replace the nibs and the ink in them. They come in an aluminum body. Uniball vision is cheap and readily available. My daily sketcher is a TWSB ECO refillable fountain pen, extra fin nib, with platinum carbon ink. You could play with fountain pens like this because based on the direction of the nib it will get a different line width and quality. This could add to the dimensions you are playing with in the designs!

I was originally imagining using something like the Pilot Razor pen. It very easily bleeds with water and makes nice grey washes. I think that they come in a few different colors and that is not the only brand of pen like that with such soluble ink.

I used a water brush pen for the all the smearing. It might be hard to keep it wet though without squeezing. Maybe the syringe can do something?

The water color pencils could also work, but it might be hard to keep constant pressure on them as the tip wears down with your XY plotter (unless I missed something for adding Z axis pressure). The Faber Castell Albrecht Durer water colored pencils are much higher quality. The pigments blend better and are more vibrant.

Last year we saw an early version of this Portrait Machine when we visited a friend in Petaluma. They layered on complexity by giving the system constraints in memory and time to mimic a real artist. The camera would look at the subject and then back at the paper with a limited memory abstraction of the subject. It would take into account what it already drew and then add to the sketch. Very philosophical and meta for what constraints to add and how much to dial them in to keep the machine output somewhat expressive. Worth a look if you ever in the area.
https://www.petaluma360.com/article/ent ... tches-art/

I really like the Jay. I agree that you are just scratching the surface in what you can do with this by playing with additional process variables. Please keep posting!

[Western Red Cedar]

I love the art and the means of production. Thanks for sharing the specifics. I look forward to seeing how it evolves. Are you building the frames yourself as well?

[black_son_of_gray]

@mF

Great suggestion to work with toned paper! I'll have to think about that next time I buy paper.

Re: inks and pens. I never realized before going into this just how many options there are in this space, and how they really do matter for specific applications beyond just e.g. writing. I've had a pretty good experience with JetPens, which is relatively local to me.

Pen/ink/paper selection is not trivial with these drawings, and I'll probably always need to do experimenting. Certain types of pens--ironically, a lot of the 'juicy' ones that are a pleasure to write with, such as Sakura Gelly Roll--simply put down too much ink and stay wet just a little too long, and maybe have something about the pigment/dye carrier that interacts with the paper in just such a way, that the saturated paper starts to fray and lift and generally degrade as the machine passes over the same spot ~a dozen or so times. End result: the paper looks and feels damaged.

The pens that have worked really well for me are Pilot V5 Extra Fine, Pigma Micron (different line widths work great!), PRO MACH 0.38, Uni-ball Signo, and OHTO graphic liners. Aesthetically, I like a high density of super-fine lines...but also, I think there might be something helpful in the line being super-fine because that means it dries almost instantly on the paper and keeps everything structurally in good shape wrt the paper fibers. From a light-fast standpoint, a lot of the black pens are permanent, archival-quality black pigments, but the colored pens are usually dyes in an alcohol carrier, and my understanding is, that they will undoubtedly fade over time. I'm not exactly sure how to navigate the decision, especially if I want to try to sell some of this eventually. If the art degrades over the course of 10 years...

Nonetheless, the idea of layering with the machine is really intriguing to me. When I was learning to sketch, I spent a few weeks trying out the simple method explained in the book "Everyday sketching & drawing : 5 steps to a unique and personal sketchbook habit" by Steven B. Reddy. The short version is this: after sketching the lines, make up a diluted india ink which lays down about a ~20-25% strength shading, then do about 3 or so successive passes with the india ink, working from light to dark, and build up the shading values on the sketch. After that, do some color washes. In essence, it's a lot simpler to do a grisaille (worry about value) separately from worrying about hue. It's quite effective! So it would be really neat to see what happens with, say, a wide linewidth pattern drawn on the machine with a e.g. light gray prismacolor marker or something else that would build on itself over successive passes. Y'now get the range of values away from binary and fill in some levels in between. I think the 'interference' pattern that emerges from that might be fascinating. But that would be a ton of wet ink, and I have my doubts about the paper...

Thanks a bunch for the test sheet you've made! I'll need to think about some of these.

@WRC

Re: frames. After making a bunch of these drawings (which have certain variables surrounding size and centering on the paper that are sometimes difficult to foresee), I wanted to frame some. Yeah, matting them would be a royal pain because it would mean each mat would need to be custom cut. So I went with a float frame. The retailer Michaels was having a ridiculous framing deal at the time, so I ended up just buying the frames from them. It was the cheapest option, even including making the frames myself, although I know now that self-made frames would have been way better build quality. Also, here's how the Michaels experience was: I bought 4 frames, went home and unwrapped them. Half(!) of them had busted corners. So I went to the store to exchange them. Opened the 2 new frames in the car in the parking lot just to be sure, turned out one had busted corners. Go back in and exchange that one, now of course the store manager wanted to see me open that package at the register before purchasing. It was fine. All in all, I had to open 7 frames to get 4 that weren't damaged. The people at Michaels were nice about the whole situation (not their fault), and I do think the drawings look nice in the frames, but it does make you wonder whether the whole pallet got dropped at some point?! How is that even a viable business model?

Another keyword for those of you reading along that are interested in this kind of stuff (particularly if you are interested in coding/algorithms), check out "generative art"

[black_son_of_gray]

Album Experiences

This is kind of a weird post. I'm looking for music recommendations for your favorite 'album experiences' - basically, albums/performances that are cohesive and vibe-y...where the whole album is composed in such a way that it is its own complete experience (distinguishing it from, say, an album 'with a lot of great songs'-there needs to be a sum to the parts, or some through-line--melodically, thematically, etc.--between songs that draws them together). Some people might refer to these as 'concept albums'. Soundtracks welcome!

The point is, these are the kind of albums where you just put the thing on and listen to the whole thing through rather than pick and choose songs.

I've got eclectic musical tastes, so I'm open to just about any genre.

Some examples:

(Genre categories very loose...)

Electronic-ish:
The Glitch Mob - Drink The Sea
Daft Punk - Interstella 5555
Daft Punk - The Tron Legacy Soundtrack
Massive Attack - (Honestly almost any mix of their stuff)

Classic/Prog/Hard Rock-ish:
Led Zeppelin IV
Pink Floyd - The Wall
Pink Floyd - Dark Side of the Moon
Muse - Absolution
Red Hot Chili Peppers - Stadium Arcadium (Mostly because RHCP songs sound so distinct...same could be said about the Beastie Boys)
Tool - Lateralus

Orchestrated:
Sergio Leone - Scores to the Dollars Trilogy
Gustav Holst - The Planets
Modest Mussorgsky - Pictures at an Exhibition
Philip Glass - Glassworks
Philip Glass - Koyaanisqatsi (Soundtrack)

Neofolk:
Heilung - Lifa

Singer/Songwriter?:
Townes Van Zandt - Live at the Whole Coffeehouse
Bob Marley and the Wailers - Legend

I'll stop there. Lemme hear about your favorites (and maybe a brief description of why you like it)!

[SouthernAlchemy]

I'm not sure I understand exactly what you are looking for, but these are some albums that I tend to listen to all the way through when I have a chance. Almost all live recordings because I do enjoy when the music comes through as a whole performance. I'll let the listener categorize.

The Clash - London Calling
Johnny Cash - At Folsom Prison
Van Morrison - Live at the Grand Opera House Belfast
Charlie Parker - Jazz at the Philharmonic, 1949
Miles Davis - In Person Friday Night at the Blackhawk
Neil Diamond - Hot August Night (no judgement until you give it a listen!!)
Taj Mahal - An Evening of Acoustic Music
10,000 Maniacs - MTV Unplugged
Aretha Franklin - Live at Fillmore West

And since you mentioned them, I'll have to say Paul's Boutique by the Beastie Boys. I haven't listened in a while, but in the late 80's I think I had that one on an endless loop. So over the top, it's ridiculous. I'll have to go see if I can dig it out.

[OutOfTheBlue]

I believe the below all fit the description!

Lou Reed - Berlin
David Bowie - Low
Robert Wyatt - Rock Bottom
Tom Waits - Bone Machine (many others could fit the bill)
Arcade Fire - Funeral // The Suburbs
Radiohead - OK Computer // Kid A
Spiritualized - Royal Albert Hall October 10 1997 (includes most songs of studio album "Ladies and Gentlemen We Are Floating in Space", but I much prefer this live version)
Lisa Germano - Geek the girl
Mazzy Starr - So Tonight That I Might See
Jeff Buckley - Grace
Black Hill and Silent Island - Tales of the Night Forest (guitars instrumental)
cLOUDDEAD - cLOUDDEAD
Amon Tobin - Supermodified
Love - Forever Changes
Keith Jarrett - Sun Bear Concerts // Lausanne-Bremen Concerts // La Scala Concert // The Koln Concert (solo piano performances)
Kyuss - Welcome to Sky Valley // And the Circus Leaves Town
Miles Davis - Pangaea // Agharta
Boredoms - Super AE
Mr. Bungle - Disco Volante

And a two French ones:

Programme - Enfer tiède
Alain Bashung - L'imprudence

[delay]

An album I enjoy as a whole rather than as separate songs is Behaviour by the synth-pop Pet Shop Boys. The album embodies the various feelings of observing human behaviour as an outsider. Like a spirit disconnected from its soul. The first song is "Being Boring", and it is. From The End of the World:

The prophets all predicted extinction
The virgin spoke in apparitions
And if it all came to pass now
You feel we'd all deserve it somehow

[ertyu]

For a cohesive "album" feel, in addition to older music that was meant to be a part of an album, I also like (1) DJ sets on youtube, e.g. Hania Rani, Boris Brejcha, and other performances for Cercle, and (2) whole live shows that were posted on youtube, e.g. In Flames, Maneskin before they "made it big", etc.

Of course, these might not vibe with your taste of music but as a category, both dj sets and live shows are deliberately designed by artists as a cohesive whole that's meant to carry the crowd through a certain experience. You might also google the name of your favorite festival, e.g. Mera Luna and Wave Gothic Treffen also often post complete live shows.

[loutfard]

- Philippe Turiot playing Bach's Goldberg variations on accordeon: spotify, youtube. Worth listening to a decent high quality audio version. The usual compression loses much of its appeal.

- Nicolaus Bruhns has a magnificent oeuvre of organ music.

- Daouda Kone's oeuvre is hilariously funny for those who speak French. "La femme de mon patron" comes to mind. Mostly happy happy music. Try it.

- David De Lange's music is wonderful for those who understand a bit of Afrikaans. His most famous recordings.

- Johannes Kerkorrel made a wonderful album Eet Kreef with "Die Gereformeerde Blues Band". Musical. Poetical. Some of this album is absolutely incredible, even if you don't understand the language or the historical context entirely.

- OK Jazz made incredible soukous / Congolese rhumba.

- Vassilis Tsitsanis was the king of rebetiko and played magnificent rebetiko. Worth listening to even if you don't understand a word of Greek.

If you liked this list, let me know. I can still add quite a bit to it...