Speaker A: Hello and welcome again, everyone, to another episode of Waiting to Be Signed, a special interview episode.

Speaker B: I'm Will.

Speaker A: I'm joined by Trinity as always, and with us today, is Piter Pasma, who is, I guess, how would you call him? Maybe like the master of code, or if it was golf, he'd be the world's greatest coder in terms of his conciseness and ability to, what, optimize, I guess would be the right term.

Speaker C: But how would his golfing be?

Speaker A: Yeah. How is your golfing, Piter?

Speaker B: Oh boy. I probably, I mean, code golfing is good, but No, probably terrible at it. Yeah.

Speaker A: That's good. I think that's a sign that you're a grounded person, probably not being too good at golf. Piter, thank you so much for taking the time to join us. It's really exciting to have you on the show. I find it hard to imagine that there would be people listening who don't know who you are or have some familiarity with you, but for the sake of completeness, maybe you can give everyone a quick introduction to yourself, your history in art and code, your path to finding NFTs, and just kind of Synopsis of your career the past couple of years?

Speaker B: So I am Piter Pasma, generative artist from the Netherlands. I started coding when I was 9 and it was already creative, but the real creative graphics started probably in the demo scene in 1997. I wrote very tiny 4 kilobyte demos there in 4,096 bytes of assembly code. And I won a few prizes with that. I always kept like a connection over the year with creative coding. I'd done a lot of just in general always working to make the machine surprise me in ways. Otherwise, I've been drawing all my life, mostly cartoons, and I was brought up with a lot of love for art by my father. Who also taught me to draw, and I think I got a lot of creativity from practicing that. At some point, I started posting my work to Instagram, getting in touch with the greater generative art community that I didn't really know actually was out there. I knew a few people, their names only, and a lot of those names are now my friends, so that's really cool. I organized January '21 and '22 and probably '23 as well, if all goes right. And yeah, so via there I got contacts with a lot of people from the generative art world. And a few months after January, the NFTs started happening. I already knew about a few NFTs a little bit. I even Heard about the rare Pepes when they were made, but I, I really didn't understand them by then. But I believe it was in December 2020 when Ringers was released. Is that about correct?

Speaker C: We'll say yes-ish.

Speaker A: We can fact-check that.

Speaker B: I think so.

Speaker C: I—

Speaker B: but because I was with my family and I remember telling them, showing them it and telling them, this is, this is a special thing.

Speaker C: Did you mint any?

Speaker B: No, no, no, not at all. I didn't understand NFTs. Dimitri is a friend of mine, and he also showed and explained me this project. And the generative part I understood, or at least somewhat. His algorithms are always very interesting and curious, or maybe simpler than they look. That's what he often says. But The NFT part, it's it's it was new to me, and it took a bit of getting used to. But there was this really special realization that see when you make generative program, usually they're at least mine were of the form that you can hit refresh a bunch of times, and you get a new image every time. It varies from artist to artist, but I usually aim to go for a about 70% good image ratio because I couldn't be bothered to sort through tens or hundreds of outputs to find the gems. And one thing that I noticed when showing people such programs was that even though there was like infinite variation that you could put a lot into, they would, after 5 minutes, walk away with a general impression of having seen 50 images with like This is sort of the thing that this program does. And then I first heard of ringers, and what that allowed was a completely new way of looking at it because every output of the program was owned by someone, and they couldn't just refresh and get a new one because they had that one. And they would look at it and think, hey, maybe that looks like a goose or a fish or something, but they will like experience every output individually instead of getting the impression of, of this whole mishmash of everything.

Speaker A: Not to derail too much, but it's kind of interesting that you brought that up because I know you've been on fxhash for a while. Like some of your drops were back in 2021 when it was still very early, when the team there decided to add the ability to cycle through preview images, like say, while something is still waiting to be open for mint. That was very controversial at the time. Well, at first it was forced, it wasn't optional. And that in particular is very controversial to artists who said, no, like what you said, I only want the ones that get minted to be like the canonical outputs. And that is what the project is. And I don't want people to be able to just sit there and spin them and kind of continue to explore.

Speaker C: Yeah.

Speaker A: And then similarly, there have been some sites, generative art sites that actually allow you to kind of keep flipping until you find the one you want and then mint that way. So you kind of get to self-curate your own output. Those have, I think, have had some varying success as well.

Speaker C: It really kills the rarity thing, I gotta say.

Speaker B: Well, that was another thing, like rarities were something I had never thought of while making my art. I mean, that, as far as I know, the whole rarity idea with some things being only a few percents of the output, it comes from the crypto world. And it's Sometimes you also have to let go as an artist, I think. Some pieces work very well with rarities, and other times maybe you don't want or didn't want rarities.

Speaker C: In my mind, that's more of a factor of there's a really big gap between the people who are interested in crypto and the people who are interested in art, with the generative art scene being where, like, that part of the Venn diagram where people are coming together. My experience is that people who are more into crypto are more like into the financial world, the gaming world, the world of maximizing everything. And so the idea of being able to chase and maximize certain characteristics is really appealing. Whereas I think when we see art created by people who are more from the art world, they were like, no rarity. It is what it is. It's all subjective to your own taste.

Speaker A: Mm-hmm.

Speaker C: And that's it. And so it's an interesting thing to bring up.

Speaker B: There is also another aspect that for at least the generative artist, I mean, what you say is correct, but I think there's like a third or in-between mode. Sometimes when you're building a program, like I said earlier, I want the machine to surprise me. I want there to be an emergence of properties. And sometimes Those things, they cannot be captured easily into features. I would have loved to have a number of a feature like the number of red pieces in a sculpture, but it would require so much calculation afterwards to just figure out how much red pieces there are. So I've thought of it, by the way, of doing it, but it would have— I was already crossing too many deadlines, but There's often when I'm in a program and I want to capture certain features, they— it's almost like you have to write a detector for the features afterwards, which is like part of the fun was that it came out automatically. So, and then you have to— it sometimes feels like you're taming a little bit your program or maybe even exposing some of the mystery that goes into it. But yeah, sometimes it also works and you can put generated titles in them, which is pretty cool.

Speaker A: Yeah. That's what you did with HyperGiraffe, right? That was a generative title. I fact-checked, Ringers was January 31st, 2021. So you were very, very close in your estimate.

Speaker B: Ah, okay. Then probably I heard about it in December and then—

Speaker A: Maybe like the work in process and stuff.

Speaker B: Mm-hmm.

Speaker C: Still didn't mint it. Oh my gosh. It's okay, it's okay. I'm just making fun. But I think what you're talking about with the features, and, you know, now that we're back on Ringers, it's interesting because, as you said, there are some like rarity things that people see into it, whether it's the duck or the goose or the fox or the frog. And that's something that's like projected by the viewer versus I think maybe some of the generative art, especially what we see on fx hash, it's like, I want 70% of these to look like a fox, 10% to look like a goose, and the rest to just look like random stuff. You know, so instead of going out and seeing what is being produced, you're trying to curate it upfront with intended results. It's interesting that you're looking at it from like the reverse way.

Speaker A: Yeah.

Speaker B: You know, like sometimes you get the idea that you got these features with 4 combinations, these with 5, and the outputs is like all the possible options times some rarity value and that's it. And I wanted to have something that like, yeah, is a bit more complex than that. So it's not really that I then came up with Sculphture or anything. It was just also like there were 2 options of a program, and I haven't even really released the other program, but it was not going to have been quite as cool as Sculphture was. So I made the right call.

Speaker C: Let's talk about that for a little bit, because for anybody who is listening only because they know about fxhash, they haven't looked at your work on Art Blocks, Could you give us a quick background on Sculpture and how it was created, how it was envisioned, and just kind of its trajectory from where it originally was?

Speaker B: Well, it's quite long. In 2015, I already started with experimenting with writing my own ray marcher, ray tracer kind of thing. It was inspired by an article written, of course, by Inigo Quiles. Who is my hero, absolutely. I look up to him tremendously. He's incredible with shaders, mathematics, and also just painting with light. I don't know if he has a painting background, but that's some really, yeah, great skill that he has. I had seen the article and I'd written a version of it in Processing, but it was in Java, so it was not on the GPU. And it took about half an hour to render something quite low resolution. I knew I had to do it in shaders, but that didn't really start happening even after quite a couple of years because at some point I got a plotter and plotters only make lines. And I actually have been making a lot of line-based art since I got a plotter. And at some point I felt like I wanted to get back into this 3D stuff again, but I didn't I didn't know how to plot it, and I guess I kind of worked out a way to make hatched 3D scenes from a ray marching thing. And I made those, I think, also near the end of 2020. Yeah, at least I started. I made a lot of hatched, so with made with lines so you could use it in the plotter. 3D scenes. And for some reason, then I got into shaders again. Oh yeah, I know why, because NFTs— you can't plot NFTs. It was, uh, part of the reason why I got the plotter is because I, I, I really like to do everything from the ground up and the hard way. So when it came to printing my art, I was wondering like, yeah, but what if the person with the big printing press, what do they add to the, to the artwork? So I wanted to get that for myself and do it with a plotter the hard way. But then the NFTs came and the digital image itself actually became the object that you could sell. This changed for me the outlook. I could like make stuff with pixels again because it didn't need to be printed. So I Got into shaders again because the GLSL shading language is really fun to write in, I think. And quickly from there out, I started yeah building a simple ray marcher just because I knew now I could do it at full speed, and the first results were pretty impressive. And at some point, I was talking to Aaron Penne. Um, what should I release for Art Blocks? The one program is, uh, that I was half considering is some of the things that I've been calling the epihyperderpflardoids, which are like a Spirograph-type figures. And the other thing was, yeah, maybe I'm gonna do a Monte Carlo path tracer kind of thing.

Speaker C: What is that for people who don't know?

Speaker B: Oh, yeah.

Speaker A: The simulation, right?

Speaker B: It simulates all the light rays. Well, not all the light rays, but a lot of them. And it averages them together. The Monte Carlo part refers to them being random. So if a light ray, if it hits a diffuse surface, like a white wall, then it actually bounces off into a random direction, not in the reflected direction like it would do on a mirror. Flat surface. And if you do that for many light rays per pixel, you get a really nice soft lighting effect that's— oh well, it's not even an effect because you're doing the physics of what photons are doing. And yeah, I decided to go for that. I also, around the same time, I had come up with an algorithm to procedurally generate shapes in 3D. Which I thought also was a very interesting problem because 2D stuff I've had been working with for a long time, and 3D is just different. There's just different stuff possible in 3D than in 2D. One of the things that I got fascinated by was that the SDFs—those are the formulas I use to define the shapes. They are relatively good at finding the intersection of 2 shapes. And the intersection is if you have like 2 spheres and you put them together, the intersection shape is like the lens shape in the middle. And that's usually an operation that's hard to do in 3D. And it's also an operation that kind of gives unexpected shapes like You get a ball, you have another ball, they're like the simplest 2 objects, and then you get a lens shape. It's immediately one step more complicated, and I started experimenting with that. So the shapes in Sculpture are like based on what you get if you have— imagine this, you've got a donut, and the donut is hollow, but it's actually got 2 shells, a double-shelled donut, and you intersect that with a double-shelled sphere. What's that even look like? I don't know, but the computer can, can do it. And those are the typical kind of shapes that you find in Sculpture. They're just like grids of shapes intersected with each other.

Speaker A: So is that kind of the secret to this? Not the simplicity, but the brevity of the code for Sculpture is working from Like simple primitives in 3D to then what you're kind of describing, it sounds like this subtraction between a torus and a sphere, double-shelled as you described, to then on the fly, you know, with a little bit of randomness, form shapes that if you were to code those shapes individually, it would be very complex to do, right? So is that, is that kind of the secret to why that code base is so slight?

Speaker B: Well, there's a lot of secrets, and I'm actually, I plan to write about all of them. It's just gonna take a while. One of the things that I already knew from the demo scene days when writing my 4 kilobyte demos, I wrote them in assembly code, and that also meant you had to write, do everything yourself. There were no libraries because they wouldn't fit in the 4 kilobyte. And what I found is that if you really go drill down to the basics of how things really You can just program them out, and the code will already be pretty small if you just generate everything. If you don't use data, but you only use code, code is really small. This is how I got into sound and music, which we talked about before the recording. Part of that four kilobyte, the last four kilobyte demo is eleven hundred bytes is a tiny software synthesizer that I wrote myself that does a sawtooth with a resonant filter, a distortion, and makes beepy sounds and drums and, well, some other stuff. And it was only 1,100 bytes. So I already knew that if I would just squish the code together, And keep hammering on it, I could get it down to pretty small. I actually kind of wanted to get it to 4 kilobytes just for the old time's sake. But well, at some point you start to get diminishing returns and 6 kilobytes was already pretty good. Scared me even at some point. Like I just, some point it just fit all on the screen and I had been working on it for months and Like that realization that half a page is everything, you know, that was pretty funny.

Speaker A: So here's a question, and maybe you've been asked this before, but why? Like, why make it smaller? What is— and then I want to tie this to some conversations that we've had with, you know, for example, in our interview with Amy, our interview with CypherD. For you, I kind of get the sense that so much of what you create The code should be appreciated as art. And that might be one of the reasons that you are driven to optimize it. And that it's like, yes, the outputs are great and look at what this code made, but like also look at what I did over here and like look at what I was able to do with so few characters. So I'm curious, like what drives you to do those optimizations? And another really great example is that little thing you put on Twitter just last week, I think it was, that you could copy paste into the Chrome browser to start generating shapes on the fly, which was really fun.

Speaker B: Yeah, HyperGiraffe is also really small, by the way. Part of the reason was in the demo scene, because it started out as doing the impossible on a machine, and there had to be a challenge. But the desktop computers already then were getting more and more powerful. If you can do ray tracing on a Commodore, and people did this, Commodore 64, it's not— doesn't look great, but—

Speaker A: Yeah.

Speaker B: They did it. And that's really impressive because it doesn't even have a divide instruction. My 4 kilobytes, they ran on Pentiums, I think, and they were much faster. So already they applied these artificial boundaries of 4 kilobytes and 64 kilobytes. Like I just said, code is really tiny. If you generate everything, 64 kilobytes is an ocean. Wow. You can't imagine what you can do in 64 kilobytes of assembly code. And it quickly dawned on me that you need a team of people. You need to have a music person, graphics person, 2 code persons maybe. And I didn't know anybody, or at least not at the skill level maybe that I required. So I did it on my own in 4 kilobytes. because that was the other alternative. And it still took me half a year also for it to make a production. So that was why I did it in the demo scene. When I did it for Sculpture, it was, well, because every byte you write onto the blockchain, because the code is on-chain, every byte costs money. And I thought, hey, that sounds like a job for size optimizing. So yeah, I got to, I had a lot of fun with that. Also, I've done a lot of weird stuff with JavaScript back in the days already, because the moment that I saw you could put scripts into web pages, well, you could do a lot of weird stuff. And sometimes in order to do— make a website do weird stuff, you only had so many characters. So you had to learn size optimizing as well, or Avoiding certain types of characters. People who know what I'm talking about probably know what I'm talking about, but that's where part of my JavaScript knowledge comes from. And there is also, and I think you just touched on that, that I, I just love code and I kind of love JavaScript. Everybody hates it, but ever since it's gotten those arrow functions, I think it's become a beautiful language that I can have a lot of creative fun in. And it's not even that I want people to like look at my code. It's also beautiful. It's part of it. It's also like a personal satisfaction thing. And maybe, I mean, that's the part of the art maybe that I don't care if people get it or not, because I know it's there. There is a thing that's called Kolmogorov complexity, and it is the minimal amount of code that you need to produce a certain string of data. So if you have a string of a billion ones, it would be— what is that? A billion— a gigabyte big. to save it if it would be 1 byte. And you could also write a program that says a billion times 1, and it would just output that, and it would be much, much smaller. But this goes a lot further. Like, you have a program, you have the digits of pi, and they're not repeating and they're random, but you can still write a program that outputs all those digits. There's some kind of magic compressibility into encoding stuff with code. And there's like, you can't do, if you just saw pi and you, the number, and you didn't know what it was, or maybe they skipped the first 1,000 decimals and you don't recognize it, you couldn't write that program. But so being able to write the shortest program for a certain output also requires a deep understanding of what that output really is. And I think that's beautiful.

Speaker C: That's really well said. Thank you. It makes me think of— I was just going through your Instagram and just kind of clicking through some of the stuff that looks really interesting. And it took me a while because a lot of the stuff on your Instagram looks really interesting. But there was one thing that came up when you were, um, it was one of your early plotting posts where you're saying that this takes so much time because I'm plotting and hatching in a single direction. So it draws a line, goes back to the start of the line. It would be so much faster if I was hatching in both directions and then my plotter will never be bored.

Speaker A: Mm-hmm.

Speaker C: And this is like, I, this is just like Einstein connecting things just because one of the Demoscene videos that you also have posted is called Never Bored from 1999. And I'm just wondering if like there's something with that efficiency, like making sure that everything is being used, like just optimized as much as possible.

Speaker B: Yeah.

Speaker C: If there's like a string there.

Speaker B: I love that. It's totally not about that, but I love the interpretation. Like, yeah, it's true that a computer where all the code is, is very optimized is never bored. It's never idling and stuff. That would be— ah man, I could have put that in there somewhere.

Speaker C: You can have it for free.

Speaker A: We can edit that out and then you can keep it as an original.

Speaker B: Great. And now it was actually about just— I was pretty young and just starting like university life and all everything came and I just, even if I had free time, I always had something to do. And that was like, uh, there was so much going on in my life that I was never bored. That was what it was really about, but I really like your interpretation.

Speaker C: But there is also that interesting parallel between the efficiency of the code and how much you can fit in. And then I think also a time efficiency with the plotting. Oftentimes when you're talking about your plottable work, which is just really beautiful, by the way, if you released that on fxhash, people would just like, I think, shit themselves.

Speaker A: I actually had that like mentally flagged. Like there are artists on fxhash who release plottable work with exportable SVGs or They'll do some kind of, you know, there'll be a 2% chance and they'll do it as a trait where if you get lucky, you'll get a free plot from them. So have you ever considered doing plottable work?

Speaker B: Well, you can plot Hyper Giraffe. They're all SVGs. You can just right-click, save, and you just have to figure out how to make the colors work and everything. But there's actually some people that contacted me about That a collector of Hyper Suraf who found a person with Poller. I was happy that I didn't have to figure it out. And they just asked me if they had my blessing to do it. And of course, because it would be really cool. I haven't heard back yet from that project. The problem with a lot of my hatching programs is that they are not long form. All the like spooky sci-fi type of threatening scenes, I just come up with them. I just start building some mathematical shapes that I think are interesting, combining formulas, and then it's a little bit like doodling because then I start to think, well, what does this look like? How can I put the camera that it is most impressive? Does there need to be extra blobs somewhere or whatever? And then I start like fleshing out the scene so it looks really cool, but That's not really a long-form process, and this was the challenge I had for sculpture: to make interesting 3D shapes in in in long form. And it's pretty hard to find more shapes that I found a few, but I've been told that they look a lot like sculpture still. So I'm not entirely sure how to turn them into long-form. Things. And there's also this bit of conflict in me that— but for new work it can be different. But for my old black and white line artwork, it was intended for the plotter, and I feel like if I turn it into an NFT, it might not— well, it is hard to have a plot and an NFT exist next to each other. The Zybotron is a work of mine on Foundation which also uses the hatching algorithm. Which I explicitly did in color, so it can't really be plotted. So it is, it is digital work, and that's why it's available as an NFT. And I'm still kind of going back and forth for how I want to approach that. But of course, if I did a long-form vector graphics thing on fx hash, then it would make sense, I guess.

Speaker C: One of the other threads that often comes up on your plotting, plottable pieces is the need to get the number of lines lower so that it's actually like efficient from a plottable perspective. If it's, you know, 30,000 lines, it's going to take forever. So how do you get that unique shape and that, that unique, I guess, aura of a piece with something that is Like a little bit more simpler, I guess, or just fewer lines altogether so it can actually be feasibly plotted.

Speaker B: Yeah, it has to do with the, um, uh, the, well, the Axidrop plotter at least. It differs if you have a like high-powered, like the old plotters from the 1980s. Um, GenGeo Merchants has one, uh, Joshua Bagley, his one goes really fast and it Doesn't really matter, but you almost feel sorry for the pen. But the servo from the AxiDraw, it's quite slow. So the moving of a pen up and down is actually quite a slow thing to do. And if you have to do that a lot, it adds a lot of time to the plot. Yeah, you just plot a lot faster if you only have one line. I mean, plots can take really, really long, like 8 hours if you want a big one. So it really makes sense to optimize them a little bit. Like some of my programs, they actually output the lines in a completely random order. And if you plot them in that order, the plotter will be not drawing 90% of the time because it'll be moving all over the page between the lines.

Speaker A: You're making the plotter intentionally bored there.

Speaker C: Which is art in and of itself.

Speaker A: It could be. Well, I'm curious, you know, we've talked a lot about sculpture and that's obviously your big Art Blocks drop. And we've talked a little bit about HyperDraft or hinted at some of your fxhash stuff. I'm curious, like, how do you decide these days which pieces you're going to submit or try to put over on Art Blocks and do on Ethereum versus which pieces you decide to publish on fxhash? And like, what's the difference between those experiences for you and Do you have a preference? Like, do you care at all? Are you chain agnostic?

Speaker B: A little bit. If it was me, I'd be chain agnostic about it. And that's the main reason why I've mostly released on fxhash after Art Blocks. Actually, only Sculpture is on Art Blocks. Part of the reason is that it just, it feels lower effort to put something on fxhash. website. You can do it all on your own. You don't need to have an appointment on the Discord with people, and it feels like you can make mistakes more.

Speaker A: For clarity, you mean lower effort in the process of publishing, not lower effort in the art?

Speaker B: Yeah, yeah, yeah, yeah. It's a smoother experience a little bit, and, and also there's not a lot of risk in the, uh, because minting is very cheap. And I also, I hadn't really thought of it. One reason to, I will probably mint some more stuff on Art Blocks in the future. And one good reason is that I also think that the on-chain part of the code is, I mean, like I said, I love the code. And if it's like hammered into a blockchain, that's extra cool, I think, although IPFS, Well, you've also released on some of the other Tezos platforms as well.

Speaker C: So a lot of work on HEN, a lot of work on Versum as well. I actually was just looking at your Wobbly Tofu piece on Versum and it's so cool.

Speaker B: Thanks. Yeah, I was going for the more like one-one pieces or like the not long-form pieces. I used to release them on Hic Et Nunc, and all sorts of stuff happened with Hic Et Nunc, and then Versum appeared, and Versum just looks really good. I just haven't really put a lot of stuff there yet, but everything about the site really pleases me, how it displays the art and facilitates things for the artist and that kind of stuff.

Speaker C: Definitely good for more of like the, the edition size pieces rather than, I guess, more of the generative.

Speaker B: I'm not even sure if you can— can you put on interactive pieces on— oh yeah, I did, I did. I have one there. That's right, the, the IMU text code poem thing.

Speaker C: Going back to fx hash, because again, we've talked about Hypergraph, We haven't really talked much about Dot Product at all, which was one of the pieces that you released before we were on the platform as well.

Speaker A: Both of them. Yeah, we joined a little, I think, after both those releases. I never got the full story, Trinity. I know you didn't get the full story. So what is kind of the genesis of both of those works, or are they connected, or is there a short version of them you can share?

Speaker B: Basically, FX has just appeared, and I decided to mint whatever the first thing I could come up with was going to be. And it was a fun blobby shader thing because I had been toying around with those on Hic Et Nunc as well, but then in the form of these click for new art. Kind of interactive pieces. So, uh, the old long-form type of thing, maybe. The name is a play on words. The dot product is a mathematical operator, but it's also the way that the different textures of the blobs— some of them are dots— they're determined in some way by the formula that Product. So it's a product. Yeah, it's the product that produces the dots, and that this made sense in my head at the time. It got copy minted after 30 minutes of posting it. That was funny by somebody called Peter Plasma, which is notorious. Also very funny because the the name Plasma is actually the name for these. Type of blobby shapes in the demo scene. We used to call them plasmas, and they were always one of my favorite effects, not just because of my last name, but just because it was, yeah, fun to just put formulas together like that. And they're really suitable for shaders.

Speaker A: Do you have like a number theory or set theory background? That's where dot product is from, right? Did you study math and stuff in School? I'm just curious.

Speaker B: It's from linear algebra. And I learned that with computational science, which I studied in university.

Speaker A: I studied that as well, but at a school that didn't give grades. So my recollection of it is, yeah.

Speaker B: I can highly, highly recommend, well, if you, linear algebra is the best thing that you will learn for math and coding. And there's a YouTube channel called 3Blue1Brown. They make a lot of educational math videos, and they have a series about linear algebra that's really, really great. And it uses a visual approach to what all these things are that you're doing. Like, who cares what a determinant is if you don't know what it's for? But once you see that it's, oh, it is like how much the space is being squeezed or not, then it suddenly makes sense, or at least a little bit. And yeah, so everyone learn linear algebra.

Speaker A: Okay.

Speaker B: Especially with projects.

Speaker A: Well, you know, to continue down the fxhash path, you've released some other work recently. Well, one that you released right before the end of beta, Angry Noise, and then one that you released shortly after, Gontian, I think. Is that— that's how I pronounce it. Does that sound right?

Speaker B: Yeah, Gonzeon is just like Hyper Giraffe, named after its random title generator. During the publication process of Hyper Giraffe, I was— you have to click through the variations to find the one that you want to keep as the COVID picture for the set, and you also see the features flipping by. So I saw one that was called Hyper Giraffe and I thought, That's a good title. And I did the same for Gonzion. I'm not so sure if it's a great title, but yeah.

Speaker C: It works.

Speaker B: I also don't know how to pronounce it because they're like made-up words.

Speaker A: Gotcha. I didn't know that about— I knew that about Hyperdraft, but I didn't know about the more recent projects. So I'm curious, you know, what— for Angry Noise, I saw a lot of work-in-progress stuff that you put editioned on some other platforms. But for both of them, I feel like I'm not quite sure what is happening on the code side or happening on like the conceptual side. And I figured now would be your— and I know that you listened to one of our shows talking about Ganzian in particular.

Speaker B: Mm-hmm.

Speaker A: And like, I felt in that show I was remarking like, I feel like there's something here and I just don't, I don't know exactly what I'm gonna get. And I would love to like take this opportunity to have you talk a little bit about those projects and get some more info out there.

Speaker B: Yeah, the thing is that I'm not good at pricing or picking edition sizes, and clearly with Gonsion I made a mistake. There was not— yeah, I'm sorry to say there was not a lot of thought about that behind it. I just— when a project nears completion, I just want to put it out there, and sometimes I think the wrong things about prices or edition sizes. So I probably have to stop the Gonzion minting process at some point. Or actually, I wanted to hear that from you because you were— I'm not sure which one of you it was, but that said, uh, I hope he's gonna keep it open for minting forever. And I was wondering why.

Speaker C: I think that At the time, you know, we hadn't had this conversation yet. There's this idea to me of if it's an artistic choice, or if it is an artistic choice to create 1,000 editions because you want to see 1,000 variations out there, then you should stick to your guns and you should do that.

Speaker B: Yeah.

Speaker A: Yeah, I think we were giving benefit of the doubt of like that there's a lot of intentionality here in the edition sizing because when we talked to other artists especially, They agonize over making sure that the edition size that they pick is like perfect for what they think their algorithm can represent. And so we were kind of carrying that forward.

Speaker B: I agonize as well. Maybe not long enough.

Speaker C: I think we could use less agony in the world, so I think it's fine.

Speaker A: Well, what amount, what amount do you think it should be now in retrospect? Or like, what would you burn it down to if you were going to do it?

Speaker B: Well, I probably would just say like, I'm gonna close it a week from now and see. But that's not an announcement, by the way. I will officially put that on Twitter also. I don't know when this will be broadcast. And, uh, but, um, I made some last-minute choices with Goncian on the designs and stuff that came out a bit different than I expected. Like the alien background stuff being very dominant in the picture. I think one of you also remarked that, and you especially see it in the gallery view. They all start kind of looking samey. So that was maybe not the best choice. On the other hand, I'm still really proud of the fractal patterns that are in there because they're— yeah, and then that was my idea behind large edition size was because there would be so many fractal patterns possible. But I didn't think it through entirely. And the other thing is that I don't think— because a lot of people, they like the bright backgrounds a lot. And I can kind of see that. And I also heard people complaining like, I minted 3 and I didn't get one. And I don't think that's cool. That would be a reason for like stopping the project to stop people from getting disappointed like that or something. And there will be more fractals in the future.

Speaker C: I think it would be interesting because obviously the, the sizing, the pricing, that is something that is obviously important that people care about. But I would like to hear more about your process. You know, going in to create Ganzian and Angry Noise and Hyperdraft, you know, we talked a lot about what your process was and some of the aspects that were important with Sculpture, especially going onto Art Blocks and, you know, making sure that it was small enough to fit into the blockchain and not be expensive. But what were some of, what was your approach when creating your fxhash releases?

Speaker B: It was different every time. So for Hypersheref, it actually— Hypersheref is size optimized. And I did it that way because I had written this very tiny quad tree algorithm, which is a spatial querying optimizing algorithm that allows you to, well, make doodly things like Hypersheref. Or any kind of stuff that requires you to see how close other points are and draw like that. And I, I had written that. And so I started looking for a use for that tiny bit of code. And then I, I had this doodling algorithm already. I had— there are some variants of it on Hic Et Nunc and on Instagram. I also posted. Some of. So I reimplemented the doodling algorithm using only very tiny code and squishing it together. And yeah, to be honest, I had not expected Hypergraph to be that big of a hit. In hindsight, I really recognized that it was like a couple of things. Like, I feel like the longer I have things brewing, the better they'll be because I keep working on them. And this was a couple of things. It was a great quad tree algorithm. Yeah. Really cool doodling algorithm and some wisdom about these wobbly functions that define the shapes of the Hyper Seraph. The brilliant insight to give all 4 corners of the Hyper Seraph a different scale. That's how they get all the different shapes. Each corner can be bigger or smaller and it interpolates between. And I actually had wanted to do the generative title thing for Sculpture. But I had wanted to make this kind of crazy science fiction backstory that is sort of that I'm now making up by myself by coming up with all these titles for the sculpture things that I put on Twitter. But the idea was to generate them. But there was a bit of a problem that, well, I just couldn't get it right. I couldn't get it right. I couldn't figure out how to make it tiny, even though it didn't have to be tiny because the features don't go on-chain. So you could actually make those big. But I didn't do it for Sculpture. And for HyperZuref, well, it was really an afterthought. I had already made the program and I thought, yeah, these random titles and Fx hash can also do it, and I felt creative because fx hash was new and there was like room to experiment. And I feel like yeah, the titles are a little bit magical realism, but bit of silly, maybe children's story kind of things. And I wanted sculpture to be more important sounding, and I couldn't figure out how to generate that. But for HyperCeref, the playful titles work very well with the images, I think. So yeah, I put them in afterwards, and I was also satisfied about being able to get them tiny. So that's for HyperCeref. For the other things on FXHash, I actually didn't size optimize it, and you can actually read the code. There's comments of me inside the code. I got some remarks about DotProduct from Alexis André, or MacTutu, I don't know how you pronounce it, that he got some use out of my comments that were in DotProduct. So I thought, hey, people actually read that. And because the size optimizing stuff, there was not— there's not really a reason to do it on fxhash except for showing off. And I already had done that a few times and that was enough. So I decided I will write nice code with jokes. Angry Noise was based on an article by Inigo Quiles, of course, which was actually about a way to create this type of noise in 3 dimensions. But the way that he did it was kind of interesting and I wanted to try it out for 2 dimensions and I did some variations on it and suddenly came out this, well, very angry noise. And I just, uh, I just wanted to showcase like the texture of that angry noise because I thought it was so special. It was something that I hadn't seen before. So that's the main intention behind it, I think.

Speaker A: You know, hearing when we had— when we talked a little bit about music before we started, And then thinking to the way that the outputs look on that project, I kind of thought maybe it was like a play on words of noise as in like noise in generative art. But also it does look like a wave pattern in a way that you, for sound, that would just is not real from physics, right? It kind of looks like a made-up wave pattern to me. So I was kind of making that connection in my mind that there might also be a reference here to sound.

Speaker B: I was also thinking I could have called it Punk Noise for, for being angry and punk, but also being a play on the word pink noise. If you've ever heard, it's a particular— sounds like a waterfall, and exactly like a waterfall. And yeah, no, it is definitely— Yeah, well, it is in 2 dimensions, so it's For me, that's— I have to look at Angry Noise now for a bit to see that if you get that waveform idea from— yeah. Or is it because there's a bar through it every time?

Speaker A: It's the bar. Yeah, the bar and the way the shapes or elements protrude from it, it just reminds me of— and often they're opposing. So, it's like a waveform going up and a waveform going down and then sometimes a third one coming back up. Yeah. To me, it was kind of like You know, there's square waves and there's sine waves. I was like, oh, this is like a whole other type of wave. And maybe there's a double entendre there between—

Speaker B: Uh, there is this strong— there was a strong idea of making a bar and putting some shapes around it. And there's also— I like waveforms. I like waveforms the way they look. You can see in my last 4K demo, if you've seen that on YouTube, called Moog. It has the waveform of the sound, uh, shown in the beginning, and that's just because I thought it looked cool. And, and I was reading a lot about, um, uh, the synthesizer, the TB-303, uh, in those days. It makes acid house, basically. It's a very famous synthesizer, and it has— well, there's a lot of mystique around it, and I don't know which parts are real, but it is said to have a really peculiar shape of both the saw and the square wave pattern that are in it. Those look also kind of pretty. They have a more pointy, angry look. But yeah, maybe that was— I've been fascinated with those patterns, so maybe they were in the back of my mind. for Angry Noise. It was generally an attempt to make like some kind of interesting design out of these shapes.

Speaker A: Yeah, I ended up getting one right before the beta ended and there was the big burn. Did you know, by the way, that that was going to be forced, the burn on the project? Were you aware when Ciphrd closed the platform?

Speaker B: I did realize it after some point. I Yeah, again, I hadn't really thought of it that much. Did you know that someone actually went through the trouble of sending all the early minters of Angry Noise, a very custom angry NFT?

Speaker A: Yes, I got one.

Speaker B: Congratulations.

Speaker A: It's in the burn wallet now, but we're at the burn address.

Speaker B: I actually spoke, was it Ozzy from FXHash? In Amsterdam, and he said that they looked at the picture and it was actually. And I looked at it also again, and it was more like a comment on the discussion on the Discord than that it was actually agreeing with those comments. I think it might have been making fun of the FUD.

Speaker A: It absolutely was. Yeah, I think it was.

Speaker B: At first when I got it, I was definitely not amused, and, uh, but after I actually kept it because, well, it's a memory.

Speaker A: But no, the— definitely the comments, the little quotes that were pulled out were very much along the lines of things that people say in especially the price discussion part of Discord. So yeah, I guess on that, I'm curious, you know, how do you regard when, you know, you have some projects that that sell out and then do, you know, amazing volume on the secondary, like Hyperdraft, and some that, well, you know, for Angry Noise, we never got to see where it would go because it was burned very ceremoniously at the end of beta. But now you have Ganzian out there. I mean, how do you kind of cope with the underlying reality of when you put something out as an NFT, it's going to be judged on economic metrics that—

Speaker B: Yeah.

Speaker A: Don't really have necessarily much bearing to you as an artist, but that's just the reality, right?

Speaker B: Yeah, I mean, I wish it wasn't. It was easier, of course, but it is the way that it is. And the most important thing is that I cannot really see— I cannot really let it affect any what I'm gonna make, you know. But I mean, that's not hard, that's easy. I don't care enough. And it's just, I think it's really hard to make sense of sometimes and to feel like, sometimes I feel a bit responsible for things and sometimes I don't. And sometimes I think, well, maybe I should make a thing with a very low mint price so people can go nuts on the secondary again. Just for the fun of it, but then there's also this thing about you have to sort of keep that price expectation a little bit, and that's really hard to estimate, I think. And I just just multiply some numbers in my head, and it doesn't always come out right.

Speaker C: I think that you shouldn't feel that type of responsibility. It's really hard to make people in this world happy, especially when people are so trained to think about things as like financialized and the work that you put in and also just the decades of experience and optimization and actually knowing assembly, which was not a thing that I knew people actually knew. So congrats on that. I think it's important to value, you know, the time that you spend. Yeah. You know, when it comes to, you know, putting something out there for people to really have fun with, uh, are you familiar with, uh, Art for Bots by, uh, RevDanCatt?

Speaker B: Yeah, yeah, yeah, yeah.

Speaker C: Yeah. Where it's so meta. It was like 0.16 Tez just to put it out there. I think there's something like fun and playful that could be done to get things out there, but like maybe as a less serious effort, if you know what I mean. It can still be serious art, but you know.

Speaker B: Yeah, I was maybe— I was thinking of, but it all went a bit wrong. I was thinking of that thing I put in the tweet to maybe put that on fx hash as well, but then make the feature be the code that you could tweet with your parameters. And I almost had that figured out, but it was about— it was too long for a tweet. I thought it was this brilliant idea because I was sure that I had seen somewhere in my wallet or a Tezos transaction that the features were on the chain recorded, but they're not. They're not. I thought that they were. And I thought that if I can just wrangle my code in there, I'll be the first on-chain FX hash project. But that all can't happen, unfortunately. So I'm now not really sure what to do with that tweet program. But that would have been a tiny fun thing. And I've been doing a secret collab. So I can't talk about that. That's what I've been spending a lot of time on. But if you kept your eyes open, you will know exactly who it is.

Speaker A: Oh, interesting. All right. That's a little bit of bait out there for everyone listening. 'Cause I feel like we, when those collabs were first announced, and I don't know if you're referring to like official FX Hash collabs or just independently, but there were a few artists who came out and kind of said like, I am working with someone, but I didn't see your name come up. So You know, you must be— I think so. Maybe you were quiet, or maybe we just missed it because there, there were a lot of other artists.

Speaker B: Yeah, I, I did tweet it, uh, but I know who it is.

Speaker A: Uh-oh. Do you wanna— well, you can throw your guess out right now, Trinity, and then, you know, Peter doesn't have to confirm it. Yeah.

Speaker C: Uh, Ippsketch.

Speaker B: Oh, I, I don't know, really.

Speaker A: Interesting.

Speaker C: I work with people who pay attention, so everything is documented. Thank you, Web2 and your centralized databases.

Speaker B: Yeah. I mean, it's not really— I thought also that it was leaked somewhere already, but—

Speaker A: I might have just missed it. Maybe I'm the one who's out of the loop here.

Speaker B: No, Ozzy told me last week to keep quiet about it again.

Speaker C: So we'll edit that out. Okay.

Speaker A: Hey, nothing was confirmed. Nothing was confirmed. So we don't even know. Peter's just playing coy with us and letting us think we know who it is.

Speaker C: But regardless, obviously, even though you were a self-proclaimed one-man show in your demoscene days, you know, there's a lot of collaboration traditionally within that environment. So have you— is this your first time collaborating with somebody since the early 2000s?

Speaker B: Yeah, it's hard. Well, no, that's not true, by the way. I did a collab and it was a lot of fun and way less hard because I did that together with Will Staal, who is also a shader wizard. And it was great fun because we had this VS Code plugin that would allow you to be 2 cursors in one file. And we had voice chat on and we were just messing around with shaders and stuff. There's some It's on my foundation, a series called Two Cursors, One File, and they're still available because we priced them at 3 Ethereum. But hey, I was thinking like it's the effort of 2 people, so it should be more.

Speaker C: It actually should be like, not twice as much, like 3 times as much. It's exponential.

Speaker A: Yeah. Depending on how well you like the person that you're collabing with.

Speaker B: So yeah, that was easy. And with the other person who may not have been named, we're doing a bit of different approach because I'm writing— but I'm giving away way too much about this project already. But it's difficult when you don't have a very similar programming style. I mean, it was not just that Will Stahl was doing shaders, but he was also specifically doing ray marchers in shaders. So we had a lot of core similar program stuff too that I understood his code when I saw it.

Speaker A: Well, Piter, I know we've gone for over an hour in here. So I think maybe if we could do one last topic or question and then maybe look to wrap it up, because you've already been really generous with your time. I'm wondering what you are most excited about in generative art right now. Maybe looking forward to the next year, 2 years, it can be with NFTs, independent of NFTs. Does the trad art world fit in anywhere? Just off the top, like, what are you looking forward to and where do you see it all going?

Speaker B: I don't know where it's all going, but I'm definitely gonna keep making art. I'm looking to do more stuff in the trad art world because that's a lot of fun and I now have opportunities there. But I'm also not going anywhere. I still want to like project my stuff onto buildings, really big ones. That, that sounds like fun, but it's kind of far removed from the type of pieces that I'm, I'm doing currently. There's a lot of ideas that I still haven't worked out. I'm really looking forward to having my taxes done. Probably also traveling to places, meeting the other generative art people. That's been— I've done that a little bit now, and it's— I want more of that.

Speaker A: Are you coming to New York City next month?

Speaker B: Well, not next month, but—

Speaker A: Isn't it next month? There's like the NFT NYC, isn't it?

Speaker B: Oh, no, no, I haven't. I'm not really planning any trips currently, but I have to see. But traveling more than I used to, which was hardly.

Speaker C: One related question, and I think this will be maybe my last one, because you mentioned, I don't know if it was on this, this recording or just before when we were chatting, you mentioned that you had just gone to Amsterdam to meet up with a lot of other generative artists. And I think what we're seeing, at least within the fxhash community, is a very large proportion of people are from the Netherlands, more so than like any other country. It's just outsized, like, population. Why do you think that might be? Is it education? Is it just— people know each other and promote each other?

Speaker B: Well, that for sure. I mean, we do that, and I feel already— I mean, but of course every country would feel like that, I guess. But yeah, when I see a Dutch person, I feel like I'm supposed to promote them a little, even though I try to make it think that I promote the good ones, but I always thought that was confirmation bias, but every time I'm on like a really big website like Reddit or something, I get the feeling that whenever I see people who aren't American, they're often Dutch. Like, there's slightly more Dutch people than there ought to be for such a tiny country on the internet. So that's— but I thought it was confirmation bias because I'm Dutch, but Maybe we're everywhere. We speak English very well even, and I think that helps. You can just dive into the English internet without having too much trouble, and I can't do that with the German internet. I don't know a lot about it. I can read some German, but writing it is very cumbersome for me. I guess education I mean, it's good. We'll find out in a couple of decades if it's still good.

Speaker A: Everyone is issued a PC and an assembly manual at age 8, and then you all just learn to code.

Speaker B: I really wonder, because there are a lot, like I noticed that there are even more than my general confirmation bias. And there's even like Dutch people who don't live in the Netherlands, but Yeah. And a lot of them I don't even know yet. So, but there's also people like Ozzy is part of the fxhash crew, I think, and he's Dutch. So maybe that attracts more Dutch people. I have no idea.

Speaker C: We'll put it all on Ozzy. He's just recruiting people to fxhash so that there could be more awesome art.

Speaker A: Yeah.

Speaker C: But I think the education piece is also important. We were, you know, when we were talking with Amy a couple of weeks ago, you know, what does it take to turn the corner and make everybody like able to appreciate and possibly produce generative art? And, you know, having more of a math, science, coding-based background where it's a part of your education rather than something that you decided to do when you were 9. That could be something that just changes the way that people think about it.

Speaker B: Yeah, definitely. I'm doing that with my volunteer work, but that's, uh, well, yeah, we're already running out of time.

Speaker A: No, no, it's okay if you want to speak on that. I— it's, it's your podcast, so, you know, if you, if you wanted to talk a little about that to wrap it up, or— I was coming up with one last question as well. We always do like 3 one last questions.

Speaker B: Oh yeah, sure. No, no way. I don't know what to say, uh, about the volunteer So yeah, you know, since you are such a—

Speaker A: your coding background is so extensive, when you look at work that you're going to collect, do you collect purely on aesthetics or do you jump into the code first and take a look and kind of like evaluate both sides of it?

Speaker B: All right. That's a great question. Yeah. I should. But I don't. The reason is that I'm not a very avid collector. So my collection is made of people that I like and people that I know will write good code or cool code. But yeah, I do look at— if I were more serious about collection, I would definitely look at the code and There have been occasions when I saw a series on fxhash and I wondered if it was maybe done with pre-calculated textures or pictures or generated or not. So I looked at, then I looked at the code to figure that out. And I personally think that fxhash could do, or do they do that already? Do they show the—

Speaker A: They just added a label. a label in 1.0 so that if it is pre-rendered images being layered, you have to label your projects now.

Speaker B: Yeah. What I wanted to see was like just the kilobyte size.

Speaker C: Oh, I don't think so.

Speaker A: But there is a cap. There's a cap, I think.

Speaker C: It's 16 megabytes, which I think it is 16 megabytes.

Speaker B: Yeah, that would be, that would be maybe too much advantage to the size optimizers.

Speaker A: That's more projects than you'll do in your entire life. 16 megabytes. That'll be all of your, your whole career fit into that.

Speaker B: There's the Omega Point by Monotau. That is a piece whose code I'm definitely going to read. I already talked to them about how the hell they did all that stuff that they did.

Speaker A: It loads so fast too. It's so wild that it just loads instantly. Yeah, yeah, yeah.

Speaker B: And I'm really looking forward to diving into that because he described some of his techniques and there's a lot of really cool tricks going on there. So that's, that's pretty cool.

Speaker A: I noticed that you minted Amulet from Stranger in the Queue this week.

Speaker B: Yeah, because I like them. I don't know if they're— but yeah, although I wish I'd gotten another one, but—

Speaker C: They're still available to mint. Yeah, they're still plenty available.

Speaker B: All right, pressure is on. Shit. Okay, what was it called? Amulet, huh? I'll look at it later. Yeah.

Speaker A: Cool. All right. I feel like that actually does wrap it now. That was a great little diversion into what we like to collect. Piter, thank you so much for coming on, being so generous with your time and just letting everyone, you know, giving everyone an opportunity to learn more about you as an artist and your history and your perspectives. It's really appreciated. We love to do these interviews. So thank you very much.

Speaker B: Yeah, thank you as well. I had a really great time. You're great, both of you.

Speaker A: Oh, thank you. You don't have to say that. Well, you can say something different off air. Cool. All right, well, that's it. That was Piter Pasma. Hope everyone enjoyed. Thanks again to Piter. Thanks to Trinity for recording on a Sunday. And yeah, we'll see you all soon. Later.

Speaker B: See ya.