jaakkoc
Re:Proton's CEO Cancels AI 'Re-Interviews'
on Saturday October 31, 2026 @03:06AM (#50313891)
That's a whole wrong PROTON you linked (some car related company?), but the real one is not that much better either:
https://www.glassdoor.com/Overview/Working-at-Proton-privacy...
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Sunday August 02, 2026 @04:26AM (#49142296)
Check the 2nd note of the blog post (best viewed with desktop)
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Sunday August 02, 2026 @04:24AM (#49142282)
I disagree - the human face is the hardest target exactly because (most) AI upscalers make up face traits and alter expressions.
The blog post you linked uses ChatGPT, which uses generative cloud AI with billions of parameters, which does indeed make up stuff.
The models that are diff'd on this blog post uses a whole different local architecture which has nothing to do with ChatGPT or Nano Banana 2, which the likes the blog post critiques. It is more rooted in older tech pre-generative era.
I do agree with the blog post, and "fool most people" -effect is especially strong when it's an old photo from a time that is not freshly in the viewers memory, and the subject no longer looks like it.
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Sunday August 02, 2026 @04:11AM (#49142215)
Thanks for the feedback, I explicitly added that now to the blog post. The wording on this comment was off.
The end goal on working on this program has been that the user shouldn't care much about what model/architecture/whatever tech is used, and should be abstracted away as much as possible. Though in a blog post like this it should be mentioned, it was hard to break away from that self-imposed mental jail :)
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Sunday August 02, 2026 @03:53AM (#49142113)
It is what it is with the title length
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Wednesday July 29, 2026 @05:53AM (#49095313)
Thanks. Yeah, I used to do the tug-and-pull of scaling, then adding noise and sharpening, scaling a little more, adding more noise, adding more sharpening, etc. to ad nauseam. And photoshop sharpening was too much always for my taste.
At scale, this tool should speed that process up a "bit". Let's say 100s of photos.
Re:I 4x'd a 367x479 stamp-sized photo through 8 upscaling models
on Tuesday July 28, 2026 @07:16AM (#49082199)
Full disclaimer: the Enlarger app is something I did.
Fuller disclaimer: this kind of a "postmark"-sized pic blowup test case is borderline unfair to every model in the comparison, mine included.
Show HN: Enlarger • A local upscaler that keeps detail instead of smoothing it
Hey everyone! I know there are gazillion alternatives out there especially with AI and all, but this nevertheless is my take on it.I created it because I was frustrated with the options I was seeing in the market. Too heavy, too complex, UX was meh, the output created was not resonating with my style of photography, they are generally glitchy or hallucinate too much, and I dislike subscription...
Re:Trump Media to sell instant access to 'market-moving' social posts
on Friday July 17, 2026 @04:54AM (#48944894)
Crazy times we live in that a president can do this.
Re:Decoy Font
on Thursday July 16, 2026 @01:09PM (#48937250)
Cool. Now do an accessible version.
(/s)
Re:Show HN: Firefox in WebAssembly
on Thursday July 16, 2026 @12:14PM (#48936538)
Seems like it was $25k worth of tokens