6 ms·

> Some modern AI art workflows often require more effort than actually illustrating using conventional media. And this blog post doesn't even get into ControlNe

> Some modern AI art workflows often require more effort than actually illustrating using conventional media. And this blog post doesn't even get into ControlNet. Indeed. Another criticism that I can definitely somewhat see the idea behind, is that the barrier to entry is very different from for example drawing. To draw, you need a pen and a paper, and you can basically start. To start with Stable Diffusion et al, you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware. One way or another, if you want to practice AI generated art, you need more money than what a pen and paper cost.

realusername 4y ago
Stable Diffusion doesn't really need powerful hardware, any graphic card will do, it will just be a bit longer. There's even ports on smartphones nowadays.
simonw 4y ago
There are plenty of traditional art mediums that require significant financial outlays to get started: oil painting, ceramics, glass blowing etc. There are plenty of free online tools for using all kinds of AI image generation techniques, and they don't require powerful hardware, just something that can browse websites or run Discord.
- adamm255 4y ago
  Plus training, lessons and inspiration. And talent. It’s like with dreams. They can be terribly intricate and detailed, but ask me to draw something creative and I’m out.
  - wincy 4y ago
    I've probably spent at least an hour a day working in Stable Diffusion and Automatic1111 since January or so. At this point I'd call it my hobby, as instead of playing video games I'm plowing time into this. And I'm definitely seeing marked improvements in my style and what I'm looking for. I'll often start with a shotgun approach and make 64-128 pictures with a basic prompt of what I'm looking for. Sometimes there's a shape and basic composition that makes me gasp it's so much better than the others. So I'll feed that into img2img or inpainting, iterate on it, tweak settings that I just sort of have an intuitive feel for what turning the knobs does, and while away for an hour or two to make it exactly what I want. There's definitely a "dreamy fantasy art style" I'm a big fan of that would have cost me an absolute fortune to commission even one image a year ago. I can't match what is coming from the top artists on Artstation (I tried a few weeks back which was humbling, some astounding work). But it's good enough that my D&D buddies are entertained and amazed that I went through and made each one of their characters. Our DM, being someone who has released creative works, was reticent and less gung-ho on AI for awhile until he decided to start playing with the AI tools (Midjourney in his case) for a new custom campaign he's running. He's suddenly able to develop novel NPC tokens for every important character we meet, the monsters are high resolution and the convenience of using Midjourney in Discord (which he already uses to coordinate our online D&D games) has been a huge boon and enhancement to how much fun and immersive our games are. A year ago this would have cost literally tens of thousands of dollars. He's a published fantasy author so prompting aka describing a scene comes naturally to him. It's been a lot of fun seeing what he's coming up with. I'm really loving the spark of creativity I've been finding in myself where the turnaround on the old tools was too long for me to not get frustrated and give up, and to see it amongst my friends, even the ones who were initially skeptical.
minimaxir 4y ago
The cost has gone way down in the last couple months. With a super-cheap T4 GPU (free in Google Colab), PyTorch 2.0, and the latest diffusers package, you can now generate batches of 9-10 images in about the same time it took to 4 images when Stable Diffusion was first released. This drastically speeds up the cherry-picking and iteration processes: https://pytorch.org/blog/accelerated-diffusers-pt-20/ https://pytorch.org/blog/accelerated-diffusers-pt-20/ Google Cloud Platform also now has preview access to L4 GPUs, which are 1.5x the cost of a T4 GPU but 3x throughput for Stable diffusion workflows (maybe more given the PyTorch 2.0 improvements for newer architectures), although I haven't tested it: https://cloud.google.com/blog/products/compute/introducing-g2-vms-with-nvidia-l4-gpus https://cloud.google.com/blog/products/compute/introducing-g...
- tough 4y ago
  We're minmaxing those costs thanks for the data
dragonwriter 4y ago
> To draw, you need a pen and a paper, and you can basically start. To start with Stable Diffusion et al, you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware A 4GB NVidia GPU (sufficient to run Stable Diffusion with the A1111 UI) is hardly “moderately powerful hardware”, and, beyond that, Stable Horde (AI Horde) exists. OTOH, a computer and internet connection are more expensive than a pencil, even if nearly ubiquitous.
smallerfish 4y ago
Stable Diffusion works fine on a CPU - on an AMD Ryzen 5700, approx 90s per image (and I believe comparable or faster on my old i7-6700). If you want to kick off a batch in the background while you work on something else, that's plenty fast. (I use: https://github.com/brycedrennan/imaginAIry https://github.com/brycedrennan/imaginAIry).
MayeulC 4y ago
> is that the barrier to entry is very different from for example drawing Thqt got me thinking. I agree, but from another perspective: the skillset is different. Traditionally, the approach to art was very bottom-up. Start with a pen and basic contouring techniques. Understanding more advanced techniques require a lot of work (perspective, shadows, etc). "AI" art generally does away with basic techniques. The emphasis is more on composing, styling. A top-down approach. "AI" artists may be able to iterate quicker by seeing "almost-finished" versions quickly (though a skilled artist can most likely imagine their work pretty well). But most of all, the tools and required skills are very different. You don't need to know a lot about machine learning, but it certainly helps. Probably pretty far from the skillset of most current artists. And people generally fear what they don't understand. And if I was an artist, I'd be at least a bit concerned about (i) it undercutting the value of my art, (ii) having to learn this alien way of doing things to remain competitive (by way of selection, artists probably enjoy their current tools. Anyway, I imagine photography was similarly upsetting in a lot of ways. It also didn't happen overnight. I also suspect we are going to see similar improvements to output quality as in early days of photography. Another similarity is with digital music (and recording/remixing before that). I wonder if we're going to see new genres emerge as a result (the equivalent of techno/electro).
- Paul-Craft 4y ago
  Your comment in particular captures it, but I can imagine a lot of the same sort of comments on this post being made about film cameras when they came out, then again about digital cameras.
  - prpl 4y ago
    Digital cameras made burst photos go from $.25+ a frame at 5FPS to effectively free with rates at 30+FPS now. That was transformative but also lead to all sorts of lamentations about lack of skill
  - rprospero 4y ago
    I remember my university photography club trying to get digital cameras banned from campus because "art only happens in the darkroom".
rob74 4y ago
Yeah, well then, please draw an image of my dog in the style of van Gogh, using pen and paper. I would say that for most of us, the more cost-effective way to get high quality artwork will still be Stable Diffusion...
GaggiX 4y ago
you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware. None of this is true, you can easily use a colab and load the models on a drive, I have always done it this way and it works perfectly. Also if you just need to start, there are plenty of free interface online to use, colab if actually you want to dive in.
efreak 4y ago
Kaggle and Google colab both provide free access to GPUs powerful enough to run stable diffusion. Alternatively, you can use other people's GPUs through Stable Horde or one of the free services online, though this comes with more limitations regarding image size and implemented extras like Lora and textual inversion and so on.

realusername 4y ago

Stable Diffusion doesn't really need powerful hardware, any graphic card will do, it will just be a bit longer. There's even ports on smartphones nowadays.

simonw 4y ago

There are plenty of traditional art mediums that require significant financial outlays to get started: oil painting, ceramics, glass blowing etc. There are plenty of free online tools for using all kinds of AI image generation techniques, and they don't require powerful hardware, just something that can browse websites or run Discord.

adamm255 4y ago
Plus training, lessons and inspiration. And talent. It’s like with dreams. They can be terribly intricate and detailed, but ask me to draw something creative and I’m out.
- wincy 4y ago
  I've probably spent at least an hour a day working in Stable Diffusion and Automatic1111 since January or so. At this point I'd call it my hobby, as instead of playing video games I'm plowing time into this. And I'm definitely seeing marked improvements in my style and what I'm looking for. I'll often start with a shotgun approach and make 64-128 pictures with a basic prompt of what I'm looking for. Sometimes there's a shape and basic composition that makes me gasp it's so much better than the others. So I'll feed that into img2img or inpainting, iterate on it, tweak settings that I just sort of have an intuitive feel for what turning the knobs does, and while away for an hour or two to make it exactly what I want. There's definitely a "dreamy fantasy art style" I'm a big fan of that would have cost me an absolute fortune to commission even one image a year ago. I can't match what is coming from the top artists on Artstation (I tried a few weeks back which was humbling, some astounding work). But it's good enough that my D&D buddies are entertained and amazed that I went through and made each one of their characters. Our DM, being someone who has released creative works, was reticent and less gung-ho on AI for awhile until he decided to start playing with the AI tools (Midjourney in his case) for a new custom campaign he's running. He's suddenly able to develop novel NPC tokens for every important character we meet, the monsters are high resolution and the convenience of using Midjourney in Discord (which he already uses to coordinate our online D&D games) has been a huge boon and enhancement to how much fun and immersive our games are. A year ago this would have cost literally tens of thousands of dollars. He's a published fantasy author so prompting aka describing a scene comes naturally to him. It's been a lot of fun seeing what he's coming up with. I'm really loving the spark of creativity I've been finding in myself where the turnaround on the old tools was too long for me to not get frustrated and give up, and to see it amongst my friends, even the ones who were initially skeptical.

minimaxir 4y ago

The cost has gone way down in the last couple months. With a super-cheap T4 GPU (free in Google Colab), PyTorch 2.0, and the latest diffusers package, you can now generate batches of 9-10 images in about the same time it took to 4 images when Stable Diffusion was first released. This drastically speeds up the cherry-picking and iteration processes: https://pytorch.org/blog/accelerated-diffusers-pt-20/ https://pytorch.org/blog/accelerated-diffusers-pt-20/ Google Cloud Platform also now has preview access to L4 GPUs, which are 1.5x the cost of a T4 GPU but 3x throughput for Stable diffusion workflows (maybe more given the PyTorch 2.0 improvements for newer architectures), although I haven't tested it: https://cloud.google.com/blog/products/compute/introducing-g2-vms-with-nvidia-l4-gpus https://cloud.google.com/blog/products/compute/introducing-g...

tough 4y ago
We're minmaxing those costs thanks for the data

dragonwriter 4y ago

> To draw, you need a pen and a paper, and you can basically start. To start with Stable Diffusion et al, you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware A 4GB NVidia GPU (sufficient to run Stable Diffusion with the A1111 UI) is hardly “moderately powerful hardware”, and, beyond that, Stable Horde (AI Horde) exists. OTOH, a computer and internet connection are more expensive than a pencil, even if nearly ubiquitous.

smallerfish 4y ago

Stable Diffusion works fine on a CPU - on an AMD Ryzen 5700, approx 90s per image (and I believe comparable or faster on my old i7-6700). If you want to kick off a batch in the background while you work on something else, that's plenty fast. (I use: https://github.com/brycedrennan/imaginAIry https://github.com/brycedrennan/imaginAIry).

MayeulC 4y ago

> is that the barrier to entry is very different from for example drawing Thqt got me thinking. I agree, but from another perspective: the skillset is different. Traditionally, the approach to art was very bottom-up. Start with a pen and basic contouring techniques. Understanding more advanced techniques require a lot of work (perspective, shadows, etc). "AI" art generally does away with basic techniques. The emphasis is more on composing, styling. A top-down approach. "AI" artists may be able to iterate quicker by seeing "almost-finished" versions quickly (though a skilled artist can most likely imagine their work pretty well). But most of all, the tools and required skills are very different. You don't need to know a lot about machine learning, but it certainly helps. Probably pretty far from the skillset of most current artists. And people generally fear what they don't understand. And if I was an artist, I'd be at least a bit concerned about (i) it undercutting the value of my art, (ii) having to learn this alien way of doing things to remain competitive (by way of selection, artists probably enjoy their current tools. Anyway, I imagine photography was similarly upsetting in a lot of ways. It also didn't happen overnight. I also suspect we are going to see similar improvements to output quality as in early days of photography. Another similarity is with digital music (and recording/remixing before that). I wonder if we're going to see new genres emerge as a result (the equivalent of techno/electro).

Paul-Craft 4y ago
Your comment in particular captures it, but I can imagine a lot of the same sort of comments on this post being made about film cameras when they came out, then again about digital cameras.
- prpl 4y ago
  Digital cameras made burst photos go from $.25+ a frame at 5FPS to effectively free with rates at 30+FPS now. That was transformative but also lead to all sorts of lamentations about lack of skill
- rprospero 4y ago
  I remember my university photography club trying to get digital cameras banned from campus because "art only happens in the darkroom".

rob74 4y ago

Yeah, well then, please draw an image of my dog in the style of van Gogh, using pen and paper. I would say that for most of us, the more cost-effective way to get high quality artwork will still be Stable Diffusion...

GaggiX 4y ago

you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware. None of this is true, you can easily use a colab and load the models on a drive, I have always done it this way and it works perfectly. Also if you just need to start, there are plenty of free interface online to use, colab if actually you want to dive in.

efreak 4y ago

Kaggle and Google colab both provide free access to GPUs powerful enough to run stable diffusion. Alternatively, you can use other people's GPUs through Stable Horde or one of the free services online, though this comes with more limitations regarding image size and implemented extras like Lora and textual inversion and so on.