News for hackers. Stuff that matters.
Re:DALL·E Now Available Without Waitlist by mrtksn on Wednesday September 28, 2022 @02:31PM (#33011487)
It is not art and art is useful(we can disagree on what's art, the age old question).

MidJourney and others are actually useful for exploration but the outputs are not because they can't spit finished deliverables to the specs. No one is paying for a picture of Mermaid eating marmalade, trending on art station, beautiful face, sharp focus, octane 8k.

They are great for exploration, it's just that I don't believe this is the killer app for these tools. We will find out what's the killer app with Stable Diffusion because with Stable Diffusion people can experiment beyond entering some prompts.

Re:DALL·E Now Available Without Waitlist by danielbln on Wednesday September 28, 2022 @03:15PM (#33011986)
Training your own likeness into the stable diffusion (via dreambooth) and then using it is absolutely hypnotic.
Re:DALL·E Now Available Without Waitlist by TulliusCicero on Wednesday September 28, 2022 @04:03PM (#33012540)
There are situations where that kind of art is useful though. People have pointed out that it could work just fine for art for card games like Magic. Probably a lot of board games too.
Re:DALL·E Now Available Without Waitlist by mrtksn on Wednesday September 28, 2022 @04:07PM (#33012583)
Sure, some people can find it useful but IMHO that's not a product, at least a good one. Consider how much more universally useful are other products like Photoshop or Blender.

I think a major problem is reproducibility and output controllability. Rolling the dice multiple time and using some of the outputs is not good enough for most applications.

Maybe this can be solved at some point but it's not at this moment. The advantage of Stable Diffusion is that it can be possible for someone to implement it, with OpenAI this feature doesn't exist and its not useful until they implement it.

Re:DALL·E Now Available Without Waitlist by TulliusCicero on Thursday September 29, 2022 @09:57PM (#33029954)
> Sure, some people can find it useful but IMHO that's not a product, at least a good one.

It's the start of a product, but it's going to keep improving. Already with inpainting and outpainting we see some new possible uses. What NovelAI (which builds on Stable Diffusion) has shown so far of their upcoming release seems impressive, though it's hard to say how much of that is cherry picking.

> Rolling the dice multiple time and using some of the outputs is not good enough for most applications.

Hmm, is that true? I feel like most of the time art that companies want is something made far ahead of the consumer seeing it, so generating a 100 versions of something and picking the best seems fine, especially if you can then use inpainting and img2img to fine-tune it.