(Photo by Jon Tyson)
This is an important episode. Probably more important than other episodes. And more urgent too. Why? You need to prepare yourself for a dramatically and very quickly changing creative landscape. And Chris is here to help you with the first steps.
- [NEWS] Affinity 2 Is Here : Affinity Photo/Designer/Publisher 2 are here. All three products are available on all platforms (Mac, Windows, iPad).
And yey, still no subscription. You can get an early-bird on their new universal license (for all products on all platforms) for $119, the full price later will be $199.
No DAM, e.g. no Lightroom “killer” yet though. Something many had hoped for.
Some of the new features are nice. New layer panel: great. Designer 2: vector warp and shape builder are real fun to play with. Photo 2: finally non-destructive RAW development (first step to the DAM?), live and compound masks and live mesh warp are really useful. Publisher 2 now lets you combine multiple docs into books, supports footnotes and autoflow.
I like what they did. I also hope they’ll keep being great competition for Adobe. In a competitive landscape we all win.
- [OTHER] : Switching over from 2 systems to one has been a bit of a game changer. Chris discusses the pros, cons and risks and how to mitigate them.
- [PHOTO, SPACE] Largest Digital Camera : Usually we try to make cameras smaller. And lighter. And more portable. Well, scientists are now building the world’s largest digital camera and it’s part of a project called the Legacy Survey of Space and Time, the LSST. It will help astronomers study billions of galaxies over the next 10 years. It’s 1.65 meters tall, has 189 sensors, has 3.2 gigapixels and its larges lens has a diameter of 1.57 meters.
- [AI, PHOTO] AI-Generated Photos of Chris : A gallery of AI-generated pictures of Chris. All it took was upload 30 selfies to a website, click a button to train a neural network on those pictures and then type in prompts to generate new imagery. Gamechanger? You bet.
- [PHOTO, AI] THE TIME IS NOW : Okay, this is important. And urgent. If you’re in the field of making art of any kind, be it photography, graphic design, painting, digital art, gaming assets, stock photography, product photography, you name it, if you’re in any of these fields, then NOW is the time to at least get your feet wet and familiarize yourself with the concepts and with the possibilities and with the limitations. Everyone needs to get at least an idea of a) what’s possible, b) how it works and you need to start thinking of how this could influence your line of work, how this could be a threat or a tool for you and how you might benefit from knowing a bit more about AI-generated art. The developments that this field is seeing right now are going really fast. So the time is NOW, not next year, not whenever.
And I’m especially talking about one new development and that is the ability to train your own network on pictures of your own person / item / thing. And that neural network will then be able to reproduce that person / item / thing in as many contexts and situations and light settings as you like. Have another look at the gallery that I shared earlier. All of these pictures are based on training a network with 30 selfies of myself. The training was fully automated, it took 45 minutes, it cost me 10 bucks and now generating a set of new selfies based on any text prompt describing what I want to generate, takes a couple of minutes and costs me a couple of cents.
Is it all perfect? No. Is it improving? Yes, and fast. The last 90 days of Stable Diffusion being in the open source have been one of the biggest game changers and liberators of creativity that I’ve ever seen in my entire life.
And I’ve seen many examples of democratizing the arts. And I’m happy to say that I’ve been part of several of these.
Let me give you a few examples.
There used to be a time when all music you heard on the radio was created in a professional studio. You needed specially treated rooms, you needed expensive microphones, you needed expensive effects gear and dynamic processors and 24-track tape machines. Then the mid 90s came along and things turned digital and you could record those 24 tracks on a PC, and you could run effects plugins that mimicked their expensive hardware brethren to a t. And many professional recording studios went under. I came from what back then was called the “home recording scene” and I ended up recording, mixing, mastering and producing music without owning an expensive studio, and that music was on par with many studio productions. And the gatekeepers had to reorient. They had to learn and evolve or stay by the wayside. I was part of that wave and I embraced it.
Then in the early 2000s photography went digital. And again, the gatekeepers were those with the expensive photo studios, the expensive medium format cameras, the expensive lighting rigs. And they looked down on the digital cameras and laughed at them. Those cameras weren’t up to par with their resolution and their color fidelity. Then the DSLRs dropped in price and the masses got access to new technology. And suddenly creativity was the limit, not access to expensive technology. The dark room went digital too and it was replaced by Photoshop and Lightroom. The analog workflow became digital. And I was part of that wave and I embraced it.
Then around 2005 radio went digital and turned into the podcast. That time the gatekeepers were the radio stations with their expensive studios and their big AM and FM transmitters and their radio licenses. And all of a sudden there were thousands and thousands of people like me who had something to talk about. And we had the internet to distribute what we talked about. And we found audiences. We found YOU. You’re listening to it right now. All of a sudden, every niche topic could find an audience. And I was was part of that wave and I embraced it.
And yes, right now we’re looking at the next big wave and it’s called AI-generated art. And AI-generated writing. And AI-generated music. Will it replace us? No, it won’t. Some of us for sure. But turns out it works best when working hand in hand with people like you and me. People who learn how to guide it, how to use it as a tool to supercharge their productivity.
So that’s why I suggest you give it a go. Get your feet wet. See what it’s about, learn the new tools, play, explore, see what’s possible, see where the limitations are, be informed and stay on top of the developments.
And I’d love to be one of your guides on this journey. Let’s learn together. Let’s explore together.
As a first simple step, go check out one of the many places where you can train your own model online for a few bucks. I suggest astria.ai ($10) or aipaintr.com ($2) – just to get your feet wet. Just to test the water. Both sites are relatively easy to use. You won’t need to be a coder or have any special skills, so that’s a simple enough testing ground. Astria.ai is the easier of the two, while aipaintr is a bit more involved, but both work really well.
Let’s explore this together. You’ll thank me later.
And by the way, if you want personal hand-holding, I’m also for hire. You can always book me over on sensei.photo for a private lesson to help you get up-to-speed. Just hit me up at sensei.photo.
<3 Feedback on this episode by Eran Pinhasov
(leave your own feedback at tfttf.com/hi)